INDEX

Explanations

phrases indicating deficiency or failure to meet expectations

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

Followed by "short" or similar words

np_acts-logits-general · gemini-2.0-flash

falling short

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-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PerformLayout

-0.51

RegressionTest

-0.51

cheiben

-0.47

parsedMessage

-0.46

IntoConstraints

-0.44

hup

-0.43

CrossOrigin

-0.43

ůměr

-0.43

 måneder

-0.42

ħħ

-0.42

POSITIVE LOGITS

 short

2.05

short

1.67

 Short

1.66

Short

1.62

 SHORT

1.58

SHORT

1.43

shy

1.28

 fhort

1.26

 corto

1.16

短

1.13

Activations Density 0.250%

phrases indicating deficiency or failure to meet expectations

Followed by "short" or similar words

falling short

No Comments

No Known Activations