INDEX

Explanations

instances of the word "first"

oai_token-act-pair · gpt-3.5-turbo

references to the word "first."

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

mbuds

-0.88

 Nadu

-0.87

 Gould

-0.74

uphem

-0.74

jong

-0.72

aths

-0.69

fact

-0.69

mund

-0.68

pers

-0.67

laws

-0.65

POSITIVE LOGITS

 thing

0.96

 baseman

0.95

 responders

0.92

 step

0.92

 glance

0.89

 glim

0.87

born

0.85

 installment

0.82

 lady

0.81

 hurdle

0.80

Activations Density 0.087%

instances of the word "first"

references to the word "first."

No Comments

No Known Activations