INDEX

Explanations

mentions of fines and penalties imposed for various infractions or non-compliance with regulations

oai_token-act-pair · gpt-3.5-turbo

references to fines and penalties for violations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

omorphic

-0.89

omorph

-0.88

groups

-0.74

NetMessage

-0.70

population

-0.70

ocaly

-0.69

ovie

-0.68

ierre

-0.68

ãĤ¨ãĥ«

-0.68

ãĤ¤ãĥĪ

-0.67

POSITIVE LOGITS

 levied

1.24

 incurred

0.94

 fines

0.92

alties

0.92

 imposed

0.88

 owed

0.82

 fined

0.81

 penalties

0.81

 penalty

0.77

($

0.77

Activations Density 0.030%

mentions of fines and penalties imposed for various infractions or non-compliance with regulations

references to fines and penalties for violations

No Comments

No Known Activations