INDEX

Explanations

repetitions of the word "this" in various contexts

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

start of turn 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 كومونز

-0.91

 queſta

-0.85

UnusedPrivate

-0.84

adaptiveStyles

-0.82

iſten

-0.80

transQ

-0.79

iſen

-0.79

ouſ

-0.79

 iſt

-0.76

parsedMessage

-0.76

POSITIVE LOGITS

0.59

 veliko

0.56

 boyunca

0.54

 ljudi

0.53

“

0.52

 lipat

0.52

 oči

0.51

 Brito

0.51

 mevcut

0.50

Activations Density 0.012%

repetitions of the word "this" in various contexts

start of turn user

No Comments

No Known Activations