INDEX

Explanations

comparisons between two similar things or situations

oai_token-act-pair · gpt-3.5-turbo

contrasting ideas or comparisons

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 barg

-0.65

 backlog

-0.64

onew

-0.63

 muster

-0.63

 helicop

-0.63

 plateau

-0.62

uly

-0.62

liga

-0.60

WOM

-0.60

mate

-0.60

POSITIVE LOGITS

same

0.94

 Same

0.86

WithNo

0.84

different

0.81

Same

0.81

minus

0.81

 identical

0.79

Different

0.78

 minus

0.78

 dstg

0.77

Activations Density 0.320%

comparisons between two similar things or situations

contrasting ideas or comparisons

No Known Activations