INDEX

Explanations

references to specific entities or individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns and significant names 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

 Guan

-0.68

rogens

-0.60

romeda

-0.60

Org

-0.57

VERTISEMENT

-0.57

 Rack

-0.56

FontSize

-0.55

laz

-0.55

 Almighty

-0.55

theless

-0.55

POSITIVE LOGITS

omore

0.76

azeera

0.72

deen

0.70

hett

0.68

astern

0.66

omi

0.66

encers

0.65

 cousins

0.65

inki

0.65

etr

0.64

Activations Density 0.417%

references to specific entities or individuals

proper nouns and significant names in various contexts

No Comments

No Known Activations