INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.72

court

-0.64

chuk

-0.63

racial

-0.63

-0.62

sports

-0.62

services

-0.61

legged

-0.61

ache

-0.61

POSITIVE LOGITS

llah

0.84

mble

0.83

iciary

0.82

 GOODMAN

0.78

 neighb

0.77

 sidx

0.76

 warr

0.76

 conduc

0.76

LECT

0.74

 describ

0.74

Activations Density 0.000%

No Known Activations

This feature has no known activations.