INDEX

Explanations

numerical figures related to events or incidents

oai_token-act-pair · gpt-3.5-turbo

numerical quantities and statistics related to incidents or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 widest

-0.68

lay

-0.64

unct

-0.64

estine

-0.64

ufact

-0.63

ophon

-0.60

ramid

-0.59

lord

-0.59

eda

-0.59

Laun

-0.58

POSITIVE LOGITS

 pounds

0.98

lbs

0.93

lbs

0.91

 guests

0.89

 yards

0.88

 minutes

0.88

 kilograms

0.87

 gallons

0.86

 volts

0.85

 grams

0.85

Activations Density 0.201%

numerical figures related to events or incidents

numerical quantities and statistics related to incidents or events

No Comments

No Known Activations