INDEX

Explanations

words related to acceptance, tolerance, and indifference

oai_token-act-pair · gpt-3.5-turbo

phrases indicating acceptance or lack of concern regarding various situations

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

hesis

-0.89

phabet

-0.88

rior

-0.83

lass

-0.82

agues

-0.72

iseum

-0.72

llan

-0.70

arcity

-0.69

rote

-0.68

heses

-0.68

POSITIVE LOGITS

 letting

1.06

 seeing

1.01

 allowing

0.98

 admitting

0.97

 accepting

0.93

 sacrificing

0.91

 swapping

0.89

 having

0.88

 giving

0.87

 granting

0.86

Activations Density 0.188%

words related to acceptance, tolerance, and indifference

phrases indicating acceptance or lack of concern regarding various situations

No Comments

No Known Activations