INDEX

Explanations

mentions of the word "then" before a name or title

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "then" in past contexts

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

toe

-0.77

chy

-0.69

 EntityItem

-0.64

 Parables

-0.64

bage

-0.63

mson

-0.62

issance

-0.61

 stalls

-0.61

 triangles

-0.60

 largeDownload

-0.60

POSITIVE LOGITS

 Yugoslavia

0.78

 Yugoslav

0.78

-'

0.75

 proceeded

0.70

ghazi

0.70

ceed

0.70

oslov

0.67

å§«

0.66

 prime

0.64

igslist

0.64

Activations Density 0.046%

mentions of the word "then" before a name or title

occurrences of the word "then" in past contexts

No Comments

No Known Activations