INDEX

Explanations

terms related to flexibility or adaptability

oai_token-act-pair · gpt-3.5-turbo

concepts and instances related to flexibility

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

 Massacre

-0.79

ocide

-0.78

Cause

-0.77

 Deaths

-0.77

ieth

-0.76

 deaths

-0.71

OWN

-0.70

 Cause

-0.69

origin

-0.68

????

-0.67

POSITIVE LOGITS

 flexible

3.22

 flexibility

2.63

flex

2.20

 Flex

2.04

 flex

1.73

 versatile

1.65

 rigid

1.44

 relaxed

1.28

 resilient

1.28

 modular

1.26

Activations Density 0.030%

terms related to flexibility or adaptability

concepts and instances related to flexibility

No Comments

No Known Activations