INDEX

Explanations

phrases containing the word "try" followed by an action

oai_token-act-pair · gpt-3.5-turbo

instances of attempts or efforts to try something new or experimental

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Deaths

-0.72

shire

-0.67

Journal

-0.67

ifa

-0.66

Klu

-0.63

ufact

-0.63

 Printed

-0.62

Scotland

-0.61

 concerns

-0.61

SOURCE

-0.61

POSITIVE LOGITS

 harder

0.93

 unsuccessfully

0.89

 hardest

0.88

unal

0.81

ocre

0.80

 experiment

0.78

 trick

0.78

 patience

0.76

icide

0.72

onz

0.69

Activations Density 0.083%

phrases containing the word "try" followed by an action

instances of attempts or efforts to try something new or experimental

No Comments

No Known Activations