INDEX

Explanations

quoted phrases within the text

oai_token-act-pair · gpt-3.5-turbo

quoted phrases within the text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 upon

-0.75

 afar

-0.69

 Rica

-0.69

 amounts

-0.68

 Nieto

-0.67

KO

-0.67

 hinges

-0.66

mim

-0.65

 viewers

-0.65

 outfielder

-0.65

POSITIVE LOGITS

classic

1.39

Golden

1.27

official

1.25

true

1.20

pure

1.16

ultimate

1.16

little

1.15

problem

1.14

traditional

1.13

technical

1.13

Activations Density 0.078%

quoted phrases within the text

quoted phrases within the text

No Comments

No Known Activations