INDEX

Explanations

terms related to essential or fundamental concepts or items

oai_token-act-pair · gpt-3.5-turbo

references to the word "basic" and its variations related to fundamental concepts or needs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

romeda

-0.81

igham

-0.79

isSpecialOrderable

-0.73

alm

-0.73

IDA

-0.70

inates

-0.70

peak

-0.68

pload

-0.66

haw

-0.65

ificant

-0.65

POSITIVE LOGITS

 necessities

1.19

 tenets

1.17

 arithmetic

1.00

 premise

0.98

 gist

0.95

 principles

0.93

lly

0.92

 decency

0.91

tenance

0.90

 outline

0.84

Activations Density 0.031%

terms related to essential or fundamental concepts or items

references to the word "basic" and its variations related to fundamental concepts or needs

No Comments

No Known Activations