INDEX

Explanations

references to persistent or ongoing issues or conditions

oai_token-act-pair · gpt-3.5-turbo

the word "lingering" and its variations, indicating a focus on prolonged effects or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ramid

-0.81

aughtered

-0.76

icient

-0.76

lisher

-0.75

portion

-0.73

©¶æ

-0.73

ħĭ

-0.69

abet

-0.68

opathic

-0.68

 Cosponsors

-0.68

POSITIVE LOGITS

 lingering

0.91

 remnants

0.89

 linger

0.86

noon

0.81

 undecided

0.73

 unresolved

0.73

luster

0.73

 dormant

0.70

aft

0.69

ipop

0.69

Activations Density 0.013%

references to persistent or ongoing issues or conditions

the word "lingering" and its variations, indicating a focus on prolonged effects or conditions

No Comments

No Known Activations