INDEX

Explanations

measurements and statistical values

oai_token-act-pair · gpt-3.5-turbo

quantifiable metrics and data related to performance or efficiency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

artisan

-0.79

older

-0.74

rower

-0.68

hement

-0.68

nik

-0.68

rog

-0.67

bard

-0.65

western

-0.65

sonian

-0.64

riot

-0.64

POSITIVE LOGITS

 percentage

1.17

 total

1.17

 calories

1.10

 Calories

1.09

GDP

1.08

 Percentage

1.05

 number

1.04

 grams

1.03

 mileage

1.02

 watts

1.00

Activations Density 0.848%

measurements and statistical values

quantifiable metrics and data related to performance or efficiency

No Comments

No Known Activations