INDEX

Explanations

phrases related to regular or daily occurrences

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate frequency or regularity of actions

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

 Torn

-0.85

 Patriarch

-0.80

enic

-0.78

asty

-0.73

 Stall

-0.69

tra

-0.68

 Twins

-0.67

ãĥĥãĥĪ

-0.66

 Sins

-0.64

icer

-0.63

POSITIVE LOGITS

 basis

1.12

iatus

0.79

 occasions

0.75

uation

0.73

VERTISEMENT

0.73

 footing

0.72

cale

0.72

,,,,,,,,

0.70

 thereof

0.69

 continuum

0.68

Activations Density 0.020%

phrases related to regular or daily occurrences

phrases that indicate frequency or regularity of actions

No Comments

No Known Activations