INDEX

Explanations

locations

oai_token-act-pair · gpt-3.5-turbo

repeated occurrences of a specific name or entity

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

NRS

-0.73

 horm

-0.70

MPG

-0.66

âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦

-0.62

 mitigating

-0.62

FI

-0.59

 sands

-0.58

oler

-0.57

 fitt

-0.57

FT

-0.57

POSITIVE LOGITS

eers

0.99

eer

0.93

erate

0.92

cheon

0.92

gha

0.91

thood

0.90

cy

0.90

tesy

0.90

gements

0.88

enos

0.87

Activations Density 0.031%

locations

repeated occurrences of a specific name or entity

No Comments

No Known Activations