INDEX

Explanations

patterns involving numerical and alphanumeric codes, suggesting a focus on identifiers or references

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

 الرياضيه

-0.84

 tartalomajánló

-0.72

 uintptr

-0.71

 estekak

-0.70

}))

-0.68

 nakalista

-0.68

 bağlantılar

-0.66

RectangleBorder

-0.66

 ModelExpression

-0.65

tidaknya

-0.65

POSITIVE LOGITS

0.57

ori

0.57

(":

0.57

blu

0.57

0.55

lat

0.54

blu

0.54

ara

0.54

cos

0.54

BER

0.53

Activations Density 0.016%

patterns involving numerical and alphanumeric codes, suggesting a focus on identifiers or references

No Comments

No Known Activations