INDEX

Explanations

instances of the word "you" followed by a personal action or situation

oai_token-act-pair · gpt-3.5-turbo

repetitions of the word "you."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Dimension

-0.74

 Sunset

-0.63

;;;;;;;;;;;;

-0.63

 Hours

-0.62

odor

-0.61

 claimants

-0.61

 Postal

-0.60

Dia

-0.60

 Speedway

-0.59

Ips

-0.59

POSITIVE LOGITS

're

1.43

've

1.20

 guys

0.99

 hear

0.98

 realize

0.90

 realise

0.87

'd

0.86

'll

0.85

 decide

0.84

 think

0.84

Activations Density 0.090%

instances of the word "you" followed by a personal action or situation

repetitions of the word "you."

No Comments

No Known Activations