INDEX

Explanations

references to modules and classes within a programming context

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

uxxxx

-0.43

󠁢

-0.42

Pit

-0.42

PIT

-0.38

ьаж

-0.38

 piped

-0.37

Pit

-0.36

pit

-0.36

 pié

-0.36

 Egal

-0.35

POSITIVE LOGITS

enderror

0.59

exitRule

0.56

providedIn

0.55

InjectAttribute

0.54

 HasFactory

0.54

 StatelessWidget

0.52

 Italij

0.52

 муніципалі

0.51

SourceChecksum

0.50

yntaxException

0.50

Activations Density 0.028%

references to modules and classes within a programming context

No Comments

No Known Activations