INDEX

Explanations

instances of the word "tolerance" or related terms

oai_token-act-pair · gpt-3.5-turbo

references to concepts of tolerance and acceptance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

prints

-0.79

tein

-0.73

Downloadha

-0.69

call

-0.68

grave

-0.67

wind

-0.64

pocket

-0.64

guard

-0.63

eu

-0.62

wer

-0.62

POSITIVE LOGITS

 tolerant

1.05

 tolerance

1.00

 toler

0.98

 tolerate

0.98

olerance

0.91

 intolerance

0.90

 intoler

0.87

 tolerated

0.81

olini

0.72

terness

0.72

Activations Density 0.022%

instances of the word "tolerance" or related terms

references to concepts of tolerance and acceptance

No Comments

No Known Activations