INDEX

Explanations

phrases related to statements or opinions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "saying."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

visible

-0.79

estern

-0.77

peg

-0.71

isible

-0.69

wn

-0.68

ocument

-0.68

ammy

-0.68

transfer

-0.65

èª

-0.64

POSITIVE LOGITS

 Pitch

0.62

 they

0.62

VK

0.60

omers

0.60

 apart

0.60

Af

0.57

 therein

0.57

 hello

0.56

 Brand

0.56

 ISPs

0.56

Activations Density 0.099%

phrases related to statements or opinions

instances of the word "saying."

No Comments

No Known Activations