INDEX

Explanations

words related to descriptions or explanations

oai_token-act-pair · gpt-3.5-turbo

terms related to choice and options, particularly in contexts involving alternatives or decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Hang

-0.62

 warm

-0.62

 weekends

-0.61

 office

-0.60

 Asian

-0.59

 logger

-0.58

 meetings

-0.58

 forums

-0.58

 Must

-0.56

 Alban

-0.56

POSITIVE LOGITS

ption

4.74

ptive

2.41

ptions

2.19

ptives

2.11

pt

1.25

pta

1.21

ptin

1.18

gypt

1.14

utical

1.08

utics

1.06

Activations Density 0.014%

words related to descriptions or explanations

terms related to choice and options, particularly in contexts involving alternatives or decisions

No Comments

No Known Activations