INDEX

Explanations

questions being asked

oai_token-act-pair · gpt-3.5-turbo

the verb "asked" and related questioning dialogue

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

Ĥ¬

-0.76

marine

-0.69

âĶ

-0.64

âĸ¬

-0.64

âĶĢâĶĢ

-0.63

pite

-0.63

ordinate

-0.61

 torches

-0.60

 Scouting

-0.60

smoking

-0.60

POSITIVE LOGITS

 rhet

1.08

 questions

0.98

ioned

0.93

 probing

0.90

naires

0.86

govtrack

0.86

 asked

0.85

him

0.81

wered

0.80

ask

0.79

Activations Density 0.042%

questions being asked

the verb "asked" and related questioning dialogue

No Comments

No Known Activations