INDEX

Explanations

words indicating approximate quantities or estimations

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NOPQRST

-0.96

 AssemblyCulture

-0.91

SerializedSize

-0.88

 Roskov

-0.85

ftagPool

-0.84

PyExc

-0.82

 '\\;'

-0.81

 Bourgoin

-0.77

 surla

-0.77

RegressionTest

-0.75

POSITIVE LOGITS

 inser

0.65

chn

0.57

Alongside

0.57

 setValue

0.54

 Dore

0.54

 Brü

0.53

Plu

0.52

 Browne

0.52

λλη

0.52

 Hauser

0.52

Activations Density 0.003%

words indicating approximate quantities or estimations

No Comments

No Known Activations