INDEX

Explanations

instances of the word "th" and its variants, indicating a focus on words associated with thematic contexts or specific details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

itter

-0.46

 conhecido

-0.45

 emot

-0.44

hasClass

-0.43

 known

-0.43

 conhecida

-0.42

fter

-0.42

iseries

-0.42

 acidosis

-0.41

\}^{

-0.41

POSITIVE LOGITS

 Roskov

0.75

OGND

0.75

出版年

0.73

Personendaten

0.71

GraphicsUnit

0.71

BeginInit

0.69

WriteLiteral

0.69

contentLoaded

0.68

CodeAttribute

0.66

Activations Density 0.385%

instances of the word "th" and its variants, indicating a focus on words associated with thematic contexts or specific details

No Comments

No Known Activations