INDEX

Explanations

references to hypothetical scenarios or situations

oai_token-act-pair · gpt-3.5-turbo

references to hypothetical concepts or scenarios

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 largeDownload

-0.92

lain

-0.86

hani

-0.83

 Colleges

-0.82

ainer

-0.78

gio

-0.76

oother

-0.75

Published

-0.74

hips

-0.72

itte

-0.70

POSITIVE LOGITS

hetically

1.00

hetical

0.90

 hypot

0.86

 hypothetical

0.82

iku

0.73

 Heller

0.69

pose

0.68

 ideal

0.67

 reflections

0.66

 mortal

0.66

Activations Density 0.009%

references to hypothetical scenarios or situations

references to hypothetical concepts or scenarios

No Comments

No Known Activations