INDEX

Explanations

instances where empathy or sympathy is expressed

oai_token-act-pair · gpt-3.5-turbo

terms related to sympathy and empathetic responses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ouver

-0.88

teen

-0.79

eor

-0.75

angular

-0.70

gradation

-0.69

growth

-0.69

 retention

-0.67

heed

-0.67

improve

-0.64

blem

-0.63

POSITIVE LOGITS

 sympath

1.12

 sympathy

0.92

 sympathetic

0.87

CTV

0.81

arity

0.73

 symp

0.72

 artery

0.71

phrine

0.71

ocations

0.71

 empath

0.71

Activations Density 0.020%

instances where empathy or sympathy is expressed

terms related to sympathy and empathetic responses

No Comments

No Known Activations