INDEX

Explanations

phrases with the word "said."

oai_token-act-pair · gpt-3.5-turbo

instances of reported speech

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

clad

-0.71

figure

-0.64

este

-0.63

emed

-0.59

intern

-0.59

Pont

-0.58

wed

-0.58

pees

-0.57

ï¸

-0.57

ãĥ¼ãĥĨ

-0.56

POSITIVE LOGITS

 goodbye

0.89

 hello

0.80

=\"

0.68

rius

0.65

 farewell

0.64

 href

0.62

 EntityItem

0.62

 msec

0.61

 sarcast

0.59

0.57

Activations Density 0.063%

phrases with the word "said."

instances of reported speech

No Comments

No Known Activations