INDEX

Explanations

instances of the word "Nora"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Ora" and its variations

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

ty

-0.89

lings

-0.82

states

-0.82

enegger

-0.80

poor

-0.80

lers

-0.76

starter

-0.76

ELL

-0.75

eenth

-0.73

ling

-0.73

POSITIVE LOGITS

velength

0.94

uthor

0.92

 Bore

0.89

izont

0.82

 Vista

0.78

emon

0.75

 bore

0.75

cffffcc

0.72

 tremend

0.69

 piercing

0.68

Activations Density 0.014%

instances of the word "Nora"

occurrences of the word "Ora" and its variations

No Comments

No Known Activations