INDEX

Explanations

references to limbs or limb-related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to limbs and amputations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Mil

-0.76

ween

-0.70

girl

-0.68

 Prescott

-0.67

advertisement

-0.67

KEY

-0.66

Pass

-0.65

Effective

-0.64

Premium

-0.64

kered

-0.63

POSITIVE LOGITS

 limb

1.11

 limbs

1.09

puter

1.08

 amput

1.03

aments

0.88

cheon

0.86

guards

0.81

 joints

0.76

alysed

0.73

stem

0.73

Activations Density 0.021%

references to limbs or limb-related terms

terms related to limbs and amputations

No Comments

No Known Activations