INDEX

Explanations

verbs related to power or control

oai_token-act-pair · gpt-3.5-turbo

references to power and control

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

akeru

-0.77

ocol

-0.62

antz

-0.59

 lapse

-0.59

nih

-0.59

chrom

-0.57

fail

-0.57

 Freddie

-0.55

otide

-0.55

 Saving

-0.55

POSITIVE LOGITS

 wielded

1.17

 wield

0.98

 wielding

0.92

Â¯

0.88

riors

0.85

 levers

0.83

hold

0.83

chairs

0.79

halla

0.79

tip

0.76

Activations Density 0.024%

verbs related to power or control

references to power and control

No Comments

No Known Activations