INDEX

Explanations

mentions of expulsion or expelling people from a group or place

oai_token-act-pair · gpt-3.5-turbo

terms related to expulsion from organizations or institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eds

-0.69

aver

-0.66

Mer

-0.65

 Needs

-0.65

anche

-0.64

aque

-0.62

ourke

-0.61

eyes

-0.60

acing

-0.59

oney

-0.59

POSITIVE LOGITS

 expelled

1.03

 expel

1.00

glomer

0.97

avorite

0.86

 expulsion

0.86

 ejected

0.82

 Lauder

0.80

 eject

0.79

ilon

0.74

 exile

0.70

Activations Density 0.005%

mentions of expulsion or expelling people from a group or place

terms related to expulsion from organizations or institutions

No Comments

No Known Activations