INDEX

Explanations

the word "eve" in various contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "eve" in relation to significant events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

BuyableInstoreAndOnline

-0.70

negie

-0.66

ateral

-0.65

umbo

-0.63

ajo

-0.60

ifles

-0.60

MAG

-0.60

tan

-0.59

affer

-0.58

ondo

-0.58

POSITIVE LOGITS

eve

1.18

rette

0.99

uary

0.95

ï¸

0.82

eve

0.81

Eve

0.79

eson

0.71

lyn

0.69

stand

0.68

eers

0.67

Activations Density 0.009%

the word "eve" in various contexts

occurrences of the word "eve" in relation to significant events

No Comments

No Known Activations