INDEX

Explanations

references to progress and steps taken in various contexts

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

"...step toward[s]" sequences

np_acts-logits-general · gemini-2.0-flash

step toward completion

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

Handled

-0.53

achusetts

-0.49

yo

-0.49

richTextPanel

-0.48

flink

-0.48

 temer

-0.47

 Hars

-0.47

 didReceive

-0.47

isticated

-0.47

Rhestr

-0.47

POSITIVE LOGITS

 toward

2.48

 towards

2.48

toward

2.22

 Toward

2.21

towards

2.20

 Towards

2.16

Towards

2.00

Toward

1.91

 hacia

1.43

TOW

1.38

Activations Density 0.272%

references to progress and steps taken in various contexts

"...step toward[s]" sequences

step toward completion

No Comments

No Known Activations