INDEX

Explanations

phrases related to actions and behaviors with a strong emphasis

oai_token-act-pair · gpt-3.5-turbo

words and phrases indicating actions or states of being in a narrative context

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

Not in Any Lists

No Comments

Negative Logits

 predec

-0.68

 confir

-0.63

 laun

-0.61

onso

-0.60

 memorable

-0.59

 referen

-0.58

oyer

-0.58

 standout

-0.57

racuse

-0.57

 grouping

-0.56

POSITIVE LOGITS

 lately

0.84

edge

0.67

DEC

0.66

acid

0.66

ebook

0.65

BOOK

0.63

ill

0.63

eries

0.62

 furiously

0.61

cakes

0.61

Activations Density 0.819%

phrases related to actions and behaviors with a strong emphasis

words and phrases indicating actions or states of being in a narrative context

No Comments

No Known Activations