INDEX

Explanations

phrases related to a specific word "Phoenicians"

oai_token-act-pair · gpt-3.5-turbo

references to the concept of ancestry or lineage

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

pole

-0.76

 Manila

-0.75

izoph

-0.73

balls

-0.67

holding

-0.67

bent

-0.66

tiny

-0.65

HAEL

-0.63

Ĥª

-0.63

mates

-0.63

POSITIVE LOGITS

vironment

1.38

zyme

1.10

esis

1.10

isance

0.94

oen

0.94

ograph

0.88

oint

0.83

onymous

0.80

isse

0.80

aed

0.80

Activations Density 0.024%

phrases related to a specific word "Phoenicians"

references to the concept of ancestry or lineage

No Comments

No Known Activations