INDEX

Explanations

references to previous content within the document

oai_token-act-pair · gpt-3.5-turbo

the repeated use of the word "above" to reference previously mentioned information

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

 Guard

-0.74

 Speedway

-0.69

 Reborn

-0.68

Mail

-0.66

 Knox

-0.66

izens

-0.64

eday

-0.64

ULTS

-0.63

 Rounds

-0.63

Ring

-0.62

POSITIVE LOGITS

mentioned

1.33

ground

1.01

 mentioned

0.89

ebin

0.86

 snippet

0.83

 equation

0.83

 paragraph

0.81

 caveats

0.80

 scenario

0.80

 quote

0.78

Activations Density 0.030%

references to previous content within the document

the repeated use of the word "above" to reference previously mentioned information

No Known Activations