INDEX

Explanations

the word "hang" and related terms within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

ChildIndex

-0.81

Efq

-0.75

 itſelf

-0.74

 purpoſe

-0.74

 Majefty

-0.73

 pleaſure

-0.69

 ―――――

-0.69

 Jefus

-0.68

 myſelf

-0.68

 Hift

-0.67

POSITIVE LOGITS

 hang

0.95

principalColumn

0.85

 Hang

0.80

Hang

0.78

hang

0.68

TAKE

0.64

 Take

0.62

 take

0.62

 TAKE

0.60

 hung

0.58

Activations Density 0.083%

the word "hang" and related terms within various contexts

No Comments

No Known Activations