INDEX

Explanations

phrases related to self-control and self-esteem

oai_token-act-pair · gpt-3.5-turbo

references to self-esteem and self-identity concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 rounds

-0.72

 spiked

-0.70

 layers

-0.69

 twists

-0.68

 towed

-0.68

 cyan

-0.67

 vertically

-0.67

 bells

-0.66

 Springs

-0.65

ply

-0.65

POSITIVE LOGITS

determination

1.66

esteem

1.58

expression

1.55

confidence

1.48

regulation

1.43

control

1.43

examination

1.41

awareness

1.40

management

1.39

interest

1.35

Activations Density 0.036%

phrases related to self-control and self-esteem

references to self-esteem and self-identity concepts

No Comments

No Known Activations