INDEX

Explanations

code imports and library references

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

import statements

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ftagPool

-0.60

PMailer

-0.59

UnusedPrivate

-0.59

ۜ

-0.55

帖最后由

-0.55

 OMITBAD

-0.53

transQ

-0.52

 تضيفلها

-0.52

GeneratedValue

-0.52

]-->

-0.51

POSITIVE LOGITS

 réfugiés

0.38

 réfugi

0.35

érique

0.33

 témoignage

0.33

paravant

0.32

mühle

0.32

 secours

0.31

lenmiş

0.31

használ

0.31

 Inscrivez

0.31

Activations Density 0.013%

code imports and library references

import statements

No Comments

No Known Activations