INDEX

Explanations

status codes and alphanumeric sequences

oai_token-act-pair · gpt-3.5-turbo

URLs or social media status updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 cabinets

-0.70

 Barkley

-0.66

 alphabet

-0.66

 Auditor

-0.65

 secretaries

-0.63

 Lansing

-0.62

 Granger

-0.62

Mex

-0.62

aukee

-0.61

 Pyramid

-0.61

POSITIVE LOGITS

ancy

0.79

ussion

0.79

0.72

ussions

0.69

åľ

0.68

pec

0.68

topic

0.68

eering

0.67

bt

0.67

åĽ

0.66

Activations Density 0.033%

status codes and alphanumeric sequences

URLs or social media status updates

No Comments

No Known Activations