INDEX

Explanations

Occurrences of words related to legal actions and infractions

oai_token-act-pair · gpt-3.5-turbo

terms related to obstruction and contempt in legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ted

-0.99

ting

-0.97

nir

-0.90

mint

-0.87

ed

-0.86

makers

-0.83

ded

-0.83

din

-0.83

anism

-0.82

rooms

-0.81

POSITIVE LOGITS

uments

1.07

uous

1.01

uously

0.93

kefeller

0.89

uity

0.82

 Creed

0.81

 Rift

0.78

atus

0.71

osity

0.69

Â¯Â¯Â¯Â¯

0.69

Activations Density 0.071%

Occurrences of words related to legal actions and infractions

terms related to obstruction and contempt in legal contexts

No Comments

No Known Activations