INDEX

Explanations

words related to assessments or evaluations of situations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

references to health-related assessments and their implications

oai_token-act-pair · gpt-4o-mini Triggered by @bot

state or condition descriptor

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 transfieras

-0.63

 Inscrivez

-0.55

ModelSerializer

-0.53

Descending

-0.52

 appart

-0.51

columbia

-0.49

 withdrew

-0.48

 Schluss

-0.47

 Compared

-0.47

âtel

-0.46

POSITIVE LOGITS

 influenced

1.33

 aided

1.32

 boosted

1.22

 assisted

1.15

 spurred

1.11

 enhanced

1.11

 facilitated

1.08

 stimulated

1.07

 affected

1.06

 impacted

1.05

Activations Density 0.453%

words related to assessments or evaluations of situations

references to health-related assessments and their implications

state or condition descriptor

No Comments

No Known Activations