INDEX

Explanations

words related to physical and emotional healing processes

oai_token-act-pair · gpt-3.5-turbo

concepts and references related to healing and recovery

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ARGET

-0.73

VP

-0.72

POL

-0.70

*/(

-0.69

legal

-0.66

nit

-0.65

reme

-0.64

 Heller

-0.64

ozy

-0.62

urity

-0.62

POSITIVE LOGITS

 heal

1.10

 heals

1.08

 healing

1.07

 healed

0.98

ilitating

0.90

 healer

0.85

 wounds

0.82

 Healing

0.79

stones

0.77

ments

0.77

Activations Density 0.007%

words related to physical and emotional healing processes

concepts and references related to healing and recovery

No Comments

No Known Activations