INDEX

Explanations

phrases indicating someone working in a particular field or profession

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a professional or occupational context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

worth

-0.73

Reason

-0.70

username

-0.69

casters

-0.66

 learners

-0.64

ULE

-0.62

Users

-0.61

hiba

-0.61

 explains

-0.59

soever

-0.59

POSITIVE LOGITS

 unison

1.25

 conjunction

1.18

 accordance

1.15

 vain

1.15

 lieu

1.13

ordinate

1.08

 tandem

1.04

clusively

0.96

 concert

0.94

animate

0.93

Activations Density 0.249%

phrases indicating someone working in a particular field or profession

phrases that indicate a professional or occupational context

No Comments

No Known Activations