INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 basin

-0.80

 lodge

-0.70

sul

-0.70

irit

-0.69

 opium

-0.68

"/

-0.66

 prefrontal

-0.66

 corridor

-0.66

 garrison

-0.66

 coun

-0.65

POSITIVE LOGITS

heid

0.87

0.84

urations

0.79

hess

0.78

soDeliveryDate

0.78

ickets

0.74

BILITIES

0.73

RESULTS

0.70

ows

0.69

esses

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.