INDEX

Explanations

proper names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the suffix "ze"

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

 glim

-0.92

acles

-0.81

inarily

-0.74

paced

-0.74

ials

-0.73

runner

-0.73

ially

-0.72

runners

-0.71

istrate

-0.71

ainment

-0.66

POSITIVE LOGITS

lda

1.22

zinski

1.19

ppelin

1.18

ÅĤ

0.93

leck

0.86

cki

0.85

zza

0.82

uble

0.81

ichen

0.81

pps

0.79

Activations Density 0.040%

proper names

occurrences of the suffix "ze"

No Comments

No Known Activations