INDEX

Explanations

references to beginning or starting points

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

icho

-0.63

nesday

-0.61

terday

-0.60

hered

-0.57

ISON

-0.57

ansas

-0.55

anism

-0.55

allel

-0.54

erson

-0.54

bred

-0.54

POSITIVE LOGITS

 sidelines

1.06

 waist

0.88

 outset

0.82

 rooft

0.79

 ropes

0.69

 barrel

0.69

 hous

0.69

 bench

0.68

 gate

0.67

 rear

0.67

Activations Density 0.123%

references to beginning or starting points

No Comments

No Known Activations