INDEX

Explanations

names or phrases related to the name "Josh."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the specific name "Josh" 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

anwhile

-0.65

ered

-0.64

 criminals

-0.64

ertodd

-0.64

tenance

-0.62

zsche

-0.61

eering

-0.61

 pigeon

-0.59

rupt

-0.59

 term

-0.59

POSITIVE LOGITS

nikov

1.14

awk

1.03

adow

0.96

ooter

0.96

ima

0.96

oke

0.90

IELD

0.89

ORTS

0.88

ooters

0.88

oot

0.86

Activations Density 0.011%

names or phrases related to the name "Josh."

occurrences of the specific name "Josh" in various contexts

No Comments

No Known Activations