INDEX

Explanations

terms related to physical or emotional healing

oai_token-act-pair · gpt-3.5-turbo

terms 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 @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

VP

-0.67

ARGET

-0.66

*/(

-0.66

urity

-0.66

legal

-0.63

coni

-0.63

pict

-0.62

POL

-0.60

Sop

-0.60

yip

-0.60

POSITIVE LOGITS

 heal

0.95

 healing

0.94

ilitating

0.93

 wounds

0.89

 heals

0.86

 healed

0.82

erest

0.80

 Healing

0.76

strap

0.75

 ashore

0.73

Activations Density 0.024%

terms related to physical or emotional healing

terms related to healing and recovery

No Comments

No Known Activations