INDEX

Explanations

terms related to termination or ending something

oai_token-act-pair · gpt-3.5-turbo

terms related to termination and endings, particularly in a contractual or medical 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

Negative Logits

acus

-0.74

rahim

-0.69

rouse

-0.69

iard

-0.69

annis

-0.65

htaking

-0.64

itsch

-0.64

esome

-0.63

itz

-0.62

hran

-0.62

POSITIVE LOGITS

 prematurely

0.85

 termin

0.83

 termination

0.83

 terminated

0.78

locked

0.77

angering

0.75

 terminate

0.75

 moratorium

0.74

 abruptly

0.74

ities

0.73

Activations Density 0.047%

terms related to termination or ending something

terms related to termination and endings, particularly in a contractual or medical context

No Known Activations