INDEX

Explanations

mentions of the human body part "head"

oai_token-act-pair · gpt-3.5-turbo

references to the head as a noun 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

PLA

-0.79

liga

-0.78

ieth

-0.74

issance

-0.67

itives

-0.65

rep

-0.64

abiding

-0.63

ovo

-0.63

ools

-0.62

ols

-0.62

POSITIVE LOGITS

 head

3.63

 heads

2.76

 Head

2.56

Head

2.23

head

2.08

 HEAD

2.02

 Heads

1.89

heads

1.61

 skull

1.59

 neck

1.54

Activations Density 0.024%

mentions of the human body part "head"

references to the head as a noun in various contexts

No Comments

No Known Activations