INDEX

Explanations

phrases related to locations or entities named "Tar"

oai_token-act-pair · gpt-3.5-turbo

references to the name "Tar" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

theless

-0.83

lihood

-0.83

mble

-0.73

shire

-0.73

vous

-0.71

ĸļ

-0.70

mable

-0.70

æĸ¹

-0.70

ĨĴ

-0.69

ETF

-0.69

POSITIVE LOGITS

iffs

1.03

aji

0.93

iff

0.92

onga

0.91

zan

0.91

 sands

0.89

idon

0.87

itans

0.84

usha

0.84

oux

0.83

Activations Density 0.011%

phrases related to locations or entities named "Tar"

references to the name "Tar" in various contexts

No Comments

No Known Activations