INDEX

Explanations

the word "leap" or related terms indicating significant advancement or action

oai_token-act-pair · gpt-3.5-turbo

instances of the word "leap" and its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Conditions

-0.79

 Locations

-0.71

 Integrity

-0.71

ello

-0.69

lies

-0.67

Lago

-0.66

 Commissioners

-0.65

 Representative

-0.64

Community

-0.64

 Removal

-0.63

POSITIVE LOGITS

 leap

3.47

 leaps

2.77

 stride

1.85

 strides

1.85

 jump

1.62

 plunge

1.55

 leaping

1.42

 leapt

1.40

 step

1.35

 jumps

1.26

Activations Density 0.022%

the word "leap" or related terms indicating significant advancement or action

instances of the word "leap" and its variations

No Comments

No Known Activations