INDEX

Explanations

exclamatory phrases expressing frustration or confusion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 vPvB

-0.82

 AspNetCore

-0.71

 незавершена

-0.69

DoubleQuotes

-0.69

Xna

-0.67

chyma

-0.66

PerformLayout

-0.64

)++;

-0.62

 дописавши

-0.62

makeText

-0.61

POSITIVE LOGITS

 hell

1.04

 heck

0.91

 Hell

0.90

 HELL

0.84

Hell

0.82

 Heck

0.76

Heck

0.69

Wtf

0.67

hell

0.65

WTF

0.64

Activations Density 0.194%

exclamatory phrases expressing frustration or confusion

No Comments

No Known Activations