INDEX

Explanations

words indicating the timeframe of an event in relation to other events

oai_token-act-pair · gpt-3.5-turbo

the conjunctions "and" and "or" used in various contexts

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

Mech

-0.64

IUM

-0.62

agy

-0.60

stocks

-0.60

Merit

-0.57

EMS

-0.57

Rail

-0.57

 Journals

-0.57

 Cald

-0.57

 Kard

-0.56

POSITIVE LOGITS

aft

1.11

during

1.04

 during

0.92

after

0.81

rogen

0.81

isode

0.81

behind

0.80

eret

0.80

 AFTER

0.76

anwhile

0.75

Activations Density 0.035%

words indicating the timeframe of an event in relation to other events

the conjunctions "and" and "or" used in various contexts

No Comments

No Known Activations