INDEX

Explanations

timestamps in the format of "hour:minute p.m."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'm' 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

bard

-0.72

 Atlantis

-0.68

 Levant

-0.67

 Bubble

-0.65

 Magical

-0.65

 Behavior

-0.64

 Prompt

-0.63

 Eleven

-0.63

EStream

-0.62

 Ancients

-0.62

POSITIVE LOGITS

obil

1.22

ichael

1.16

ormon

1.06

obiles

1.06

igr

1.05

umbai

1.03

otor

1.02

igrants

0.96

asonry

0.96

aintain

0.95

Activations Density 0.056%

timestamps in the format of "hour:minute p.m."

occurrences of the letter 'm' in various contexts

No Comments

No Known Activations