INDEX

Explanations

instances of speculation or uncertainty in discussion

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

educated guesses and predictions

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

StoryboardSegue

-0.56

 ModelExpression

-0.52

 direttore

-0.47

applicazione

-0.46

 morire

-0.45

 protoimpl

-0.45

 useCallback

-0.45

ActionCreators

-0.45

 gehad

-0.45

cciale

-0.44

POSITIVE LOGITS

IUrlHelper

0.48

 speculate

0.48

 interpreting

0.47

 unsub

0.47

 speculated

0.46

 interpretation

0.44

 decipher

0.44

 guess

0.43

 speculation

0.42

 guesses

0.42

Activations Density 0.113%

instances of speculation or uncertainty in discussion

educated guesses and predictions

No Comments

No Known Activations