INDEX

Explanations

verbs related to continuing or lasting

oai_token-act-pair · gpt-3.5-turbo

terms related to continuity or ongoing situations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aptic

-0.77

hedral

-0.69

 GEAR

-0.68

 Toll

-0.65

ussen

-0.65

 sights

-0.63

foot

-0.63

 radius

-0.62

 Springs

-0.60

feet

-0.60

POSITIVE LOGITS

ently

1.00

enza

0.90

rences

0.86

enance

0.84

ously

0.82

 unchanged

0.81

ency

0.78

 persisted

0.77

ichen

0.77

ences

0.76

Activations Density 0.018%

verbs related to continuing or lasting

terms related to continuity or ongoing situations

No Comments

No Known Activations