INDEX

Explanations

informal expressions indicating a sense of urgency or necessity

oai_token-act-pair · gpt-3.5-turbo

expressions of obligation or necessity

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

edly

-0.78

furt

-0.77

lies

-0.76

livious

-0.75

croft

-0.73

liness

-0.71

lund

-0.70

lier

-0.69

hips

-0.69

lique

-0.67

POSITIVE LOGITS

 gotta

0.88

nab

0.81

 buckle

0.78

avorite

0.74

 wanna

0.74

ta

0.74

FINE

0.74

 notch

0.73

 stay

0.71

 ditch

0.70

Activations Density 0.022%

informal expressions indicating a sense of urgency or necessity

expressions of obligation or necessity

No Comments

No Known Activations