INDEX

Explanations

locations or starting points in a series of potentially related events

oai_token-act-pair · gpt-3.5-turbo

recurring phrases indicating location or direction

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

rongh

-0.69

buster

-0.65

conn

-0.61

Barn

-0.61

 Mant

-0.61

Sac

-0.60

aah

-0.59

 Stacy

-0.59

Ca

-0.59

HOU

-0.58

POSITIVE LOGITS

 onwards

1.10

 onward

0.97

idate

0.81

ileaks

0.81

forth

0.76

iral

0.74

annis

0.72

awaru

0.71

uther

0.70

forward

0.69

Activations Density 0.035%

locations or starting points in a series of potentially related events

recurring phrases indicating location or direction

No Comments

No Known Activations