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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 darts

-0.71

HF

-0.69

)))

-0.68

ģ«

-0.67

Zy

-0.66

Dow

-0.65

 diamonds

-0.64

 Twitch

-0.64

LLOW

-0.64

NHL

-0.64

POSITIVE LOGITS

elta

0.77

irlf

0.76

lishes

0.74

ioch

0.73

ainment

0.73

irming

0.72

ynt

0.72

amera

0.72

atem

0.71

imar

0.71

Activations Density 0.000%

No Known Activations

This feature has no known activations.