INDEX

Explanations

words related to medical conditions induced by external factors

oai_token-act-pair · gpt-3.5-turbo

instances of the word "induced" in various contexts related to causes and effects

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

 trusted

-0.69

 name

-0.68

 piece

-0.67

 champions

-0.66

 stand

-0.65

 branch

-0.65

 live

-0.64

 alphabet

-0.64

 trust

-0.63

 buckle

-0.61

POSITIVE LOGITS

induced

3.62

 induced

2.03

mediated

1.87

imposed

1.56

inducing

1.40

assisted

1.35

 induces

1.35

 inducing

1.30

associated

1.25

related

1.21

Activations Density 0.020%

words related to medical conditions induced by external factors

instances of the word "induced" in various contexts related to causes and effects

No Comments

No Known Activations