INDEX

Explanations

terms related to automatic actions or devices

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "automatic" and its variations in different contexts

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

borough

-0.90

apest

-0.83

ĸļ

-0.82

usky

-0.79

rug

-0.74

boro

-0.72

orks

-0.72

mond

-0.71

ä»£

-0.71

 Loft

-0.71

POSITIVE LOGITS

 transmission

1.25

 induction

0.96

 transmissions

0.93

 confir

0.92

 automatic

0.92

 captcha

0.91

 braking

0.90

 renewal

0.90

 termination

0.88

 override

0.87

Activations Density 0.012%

terms related to automatic actions or devices

occurrences of the word "automatic" and its variations in different contexts

No Comments

No Known Activations