INDEX

Explanations

instances of the word "flex" and concentrations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

ſelves

-0.76

 ſou

-0.73

 ainfi

-0.72

 quæ

-0.69

 desmotivaciones

-0.68

 Conſ

-0.68

RenderAtEndOf

-0.67

ſelf

-0.65

 ſch

-0.62

 Perſ

-0.62

POSITIVE LOGITS

hy

0.72

Hy

0.72

Hy

0.66

uch

0.64

HY

0.61

flex

0.55

 flex

0.54

 concentrated

0.54

HY

0.54

hy

0.54

Activations Density 0.341%

instances of the word "flex" and concentrations in various contexts

No Comments

No Known Activations