INDEX

Explanations

patterns of characters that do not form coherent words or phrases

oai_token-act-pair · gpt-3.5-turbo

references to a specific symbol or character

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

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

raints

-0.97

matic

-0.80

 Instr

-0.75

 slic

-0.74

 Appalach

-0.72

utra

-0.71

urated

-0.68

ngth

-0.67

 Kodi

-0.67

 primates

-0.67

POSITIVE LOGITS

âĶĢâĶĢ

1.22

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

0.93

ľ

0.92

Ĺ

0.90

à©

0.90

ishable

0.89

Ķ

0.88

Ł

0.87

ĺ

0.85

×ª

0.85

Activations Density 0.042%

patterns of characters that do not form coherent words or phrases

references to a specific symbol or character

No Comments

No Known Activations