INDEX

Explanations

specific references to dictionaries

oai_token-act-pair · gpt-3.5-turbo

references to dictionaries and encyclopedias

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 SOLD

-0.79

arters

-0.71

hips

-0.71

 financing

-0.67

Fig

-0.65

fitted

-0.65

hops

-0.63

 Opportun

-0.62

 shoots

-0.62

itent

-0.61

POSITIVE LOGITS

 Dictionary

4.02

 dictionary

3.73

 diction

2.64

ictionary

2.55

 Encyclopedia

1.91

 encyclopedia

1.67

 vocabulary

1.59

lex

1.54

cyclopedia

1.54

 dict

1.36

Activations Density 0.023%

specific references to dictionaries

references to dictionaries and encyclopedias

No Comments

No Known Activations