INDEX

Explanations

terms related to leadership positions

oai_token-act-pair · gpt-3.5-turbo

references to leadership positions or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

eros

-0.73

 mell

-0.65

ĸļ

-0.64

aneously

-0.63

icity

-0.60

 Strat

-0.59

selves

-0.59

PsyNetMessage

-0.59

 luckily

-0.58

 subp

-0.58

POSITIVE LOGITS

quarter

1.15

lining

1.10

canon

0.98

gear

0.97

scar

0.95

liner

0.93

hun

0.92

phones

0.88

strong

0.87

pin

0.87

Activations Density 0.025%

terms related to leadership positions

references to leadership positions or titles

No Comments

No Known Activations