INDEX

Explanations

expressions of ability or possibility

oai_token-act-pair · gpt-3.5-turbo

phrases expressing capability or possibility

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

 hurled

-0.58

Sod

-0.56

 Booth

-0.56

ModLoader

-0.55

IB

-0.54

 Fern

-0.54

 tram

-0.54

 Trin

-0.54

 clen

-0.53

sbm

-0.53

POSITIVE LOGITS

't

0.87

eret

0.83

 afford

0.77

reach

0.76

 Provide

0.76

nery

0.76

 Improve

0.74

 establish

0.73

 succeed

0.73

oust

0.73

Activations Density 0.099%

expressions of ability or possibility

phrases expressing capability or possibility

No Comments

No Known Activations