INDEX

Explanations

references to modifications and changes made to code or content

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

making changes and modifications

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

 ujednoznacz

-0.59

 defaultstate

-0.52

haikusbot

-0.48

 Drapeau

-0.48

 Dispenser

-0.47

Peers

-0.46

 çift

-0.45

 jsPsych

-0.45

DoubleQuotes

-0.44

 camiset

-0.44

POSITIVE LOGITS

 modifica

0.44

indd

0.42

 modification

0.38

Edited

0.35

 modify

0.35

 modificación

0.35

修改

0.35

 alteration

0.34

 edited

0.33

 changes

0.33

Activations Density 0.047%

references to modifications and changes made to code or content

making changes and modifications

No Comments

No Known Activations