INDEX

Explanations

mentions of internet activity or technology

oai_token-act-pair · gpt-3.5-turbo

variations of the word "int"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

è¦ļéĨĴ

-0.87

gypt

-0.77

mares

-0.76

¥µ

-0.74

jriwal

-0.73

OHN

-0.71

¶ħ

-0.70

 tremend

-0.69

 corrid

-0.68

POSITIVE LOGITS

ellect

1.05

elligent

0.88

illation

0.84

uitive

0.81

imal

0.81

ensity

0.78

ense

0.78

ention

0.77

ended

0.77

ypes

0.76

Activations Density 0.013%

mentions of internet activity or technology

variations of the word "int"

No Comments

No Known Activations