INDEX

Explanations

phrases related to neurons and their activities

oai_token-act-pair · gpt-3.5-turbo

references to neurons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Columbus

-0.78

 Akron

-0.71

 Mankind

-0.70

 Petty

-0.67

nant

-0.64

ress

-0.64

fare

-0.62

EMP

-0.62

 Indy

-0.61

 Izan

-0.60

POSITIVE LOGITS

 neurons

1.24

 neuron

1.21

apses

0.99

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.88

cells

0.87

ergic

0.87

 receptors

0.83

 cells

0.83

 neuronal

0.82

wana

0.80

Activations Density 0.010%

phrases related to neurons and their activities

references to neurons

No Comments

No Known Activations