INDEX

Explanations

words related to skepticism and questioning

oai_token-act-pair · gpt-3.5-turbo

phrases indicating types or categories of things

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

omers

-0.95

Fs

-0.91

aries

-0.90

arks

-0.88

aughs

-0.88

anders

-0.87

essors

-0.87

Ds

-0.87

Cs

-0.87

hers

-0.86

POSITIVE LOGITS

 consolation

0.94

 miracle

0.93

 miraculous

0.92

 contingency

0.89

 mythical

0.88

 rudimentary

0.86

 unofficial

0.86

 intermediary

0.85

 sleeper

0.85

 afterlife

0.85

Activations Density 0.058%

words related to skepticism and questioning

phrases indicating types or categories of things

No Comments

No Known Activations