INDEX

Explanations

phrases related to episodes or segments of a series or show

oai_token-act-pair · gpt-3.5-turbo

unique formatting or special characters typically used in titles or references

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

FUL

-0.81

LESS

-0.71

 humming

-0.66

 unauthorized

-0.66

 referen

-0.66

hips

-0.65

////////////////

-0.63

ModLoader

-0.62

CLASSIFIED

-0.62

UGH

-0.61

POSITIVE LOGITS

iphany

1.57

hemer

1.44

isodes

1.38

iscopal

1.27

istle

1.18

onymous

1.13

oleon

1.12

alm

1.07

idem

1.07

ublic

1.04

Activations Density 0.043%

phrases related to episodes or segments of a series or show

unique formatting or special characters typically used in titles or references

No Comments

No Known Activations