INDEX

Explanations

scores or ratings within a text

oai_token-act-pair · gpt-3.5-turbo

references to scoring or ratings in various contexts

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

 detachment

-0.64

xon

-0.63

ĸļ

-0.59

etheless

-0.59

 Wonderland

-0.59

 Workers

-0.58

 Cellular

-0.58

 fascination

-0.58

 compulsion

-0.58

agan

-0.57

POSITIVE LOGITS

card

1.39

cards

1.25

ific

1.06

keeper

0.99

keepers

0.94

heet

0.87

rs

0.86

keeping

0.86

emi

0.82

ifications

0.81

Activations Density 0.032%

scores or ratings within a text

references to scoring or ratings in various contexts

No Comments

No Known Activations