INDEX

Explanations

contextual references and rhetorical questions related to user experiences and feelings

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

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

 itſelf

-0.54

 pungkas

-0.47

Filmografie

-0.46

Derbyniad

-0.45

AFFIRMED

-0.45

 henvisninger

-0.44

 myſelf

-0.44

 مرئيه

-0.44

 Monfieur

-0.43

ADELPHIA

-0.43

POSITIVE LOGITS

MessageOf

0.52

prepareStatement

0.43

gium

0.42

sizeCache

0.42

contentLoaded

0.42

Glob

0.41

 للاسماء

0.40

 proverb

0.40

 gouvern

0.39

Pattern

0.39

Activations Density 0.176%

contextual references and rhetorical questions related to user experiences and feelings

No Comments

No Known Activations