INDEX

Explanations

words indicating theories, beliefs, or ideas such as "assumption," "conclusion," or "observation."

oai_token-act-pair · gpt-3.5-turbo

terms relating to assumptions and conclusions

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

endars

-0.95

NetMessage

-0.89

inia

-0.78

cler

-0.72

pleting

-0.72

artney

-0.71

ensis

-0.70

rha

-0.70

events

-0.68

OTOS

-0.68

POSITIVE LOGITS

 uttered

0.84

 echoed

0.83

 expressed

0.81

 assumption

0.80

 glean

0.80

 about

0.79

 regarding

0.79

 premise

0.78

 articulated

0.78

 voiced

0.78

Activations Density 0.245%

words indicating theories, beliefs, or ideas such as "assumption," "conclusion," or "observation."

terms relating to assumptions and conclusions

No Comments

No Known Activations