INDEX

Explanations

phrases with the word "opposed" at different levels of certainty or emphasis

oai_token-act-pair · gpt-3.5-turbo

words related to conditions or states of being that are 'posed' or suggested

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Footnote

-0.74

istics

-0.74

adier

-0.65

 Ellis

-0.63

Jou

-0.60

 Kingston

-0.60

 commemor

-0.60

Forest

-0.59

 Staples

-0.59

Tat

-0.58

POSITIVE LOGITS

osed

1.37

mble

0.97

 conflic

0.95

 satell

0.93

itled

0.84

etheless

0.84

icter

0.82

osing

0.80

escription

0.80

pose

0.80

Activations Density 0.012%

phrases with the word "opposed" at different levels of certainty or emphasis

words related to conditions or states of being that are 'posed' or suggested

No Comments

No Known Activations