INDEX

Explanations

cooking and preparation instructions

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

cut before user

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

EndGlobalSection

-0.53

########.

-0.49

MarshalTo

-0.48

 fieldNum

-0.46

borderSide

-0.46

bootstrapcdn

-0.45

olerance

-0.45

 getCode

-0.44

SBATCH

-0.44

⟬

-0.44

POSITIVE LOGITS

énario

0.36

 sauce

0.36

Gru

0.35

CloseOperation

0.35

 solution

0.35

basicConfig

0.35

glTex

0.35

bit

0.34

RectangleBorder

0.32

 remaining

0.32

Activations Density 0.016%

cooking and preparation instructions

cut before user

No Comments

No Known Activations