INDEX

Explanations

adjectives or verbs related to something being distorted or misshapen

oai_token-act-pair · gpt-3.5-turbo

words related to distortion or manipulation of concepts or objects

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

alty

-0.85

ILA

-0.76

worthiness

-0.76

ciation

-0.75

ervation

-0.73

Â¯Â¯Â¯Â¯

-0.72

iphate

-0.72

particip

-0.71

upon

-0.70

Han

-0.69

POSITIVE LOGITS

 twisted

1.17

 twist

1.12

 twisting

1.06

 twists

1.00

 Twisted

0.86

 lengths

0.81

 intrins

0.78

ulously

0.78

 adolesc

0.74

endish

0.74

Activations Density 0.006%

adjectives or verbs related to something being distorted or misshapen

words related to distortion or manipulation of concepts or objects

No Comments

No Known Activations