INDEX

Explanations

references to fleets of vehicles or ships

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "fleet."

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

pheus

-0.75

Downloadha

-0.73

ividual

-0.71

 Hussein

-0.70

heimer

-0.69

 Extrem

-0.69

 Malik

-0.67

cond

-0.66

nai

-0.66

phrine

-0.65

POSITIVE LOGITS

 fleet

0.82

 fleets

0.80

wide

0.78

mates

0.77

 deployments

0.76

heed

0.75

 mates

0.73

 Fleet

0.72

ingly

0.71

 parked

0.71

Activations Density 0.017%

references to fleets of vehicles or ships

occurrences of the word "fleet."

No Comments

No Known Activations