INDEX

Explanations

phrases related to the technical generation or production of outputs

oai_token-act-pair · gpt-3.5-turbo

references to "output" in a technical or computational context

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

creen

-0.67

 Blanc

-0.64

 Antiqu

-0.64

Fort

-0.63

llan

-0.62

issy

-0.60

 antiqu

-0.59

anova

-0.59

vette

-0.59

ETA

-0.58

POSITIVE LOGITS

ted

1.03

 output

1.02

 outputs

1.00

 Output

0.95

ting

0.93

output

0.83

flows

0.78

rious

0.77

flow

0.76

 overflow

0.75

Activations Density 0.010%

phrases related to the technical generation or production of outputs

references to "output" in a technical or computational context

No Comments

No Known Activations