INDEX

Explanations

phrases indicating instructions or guidelines

oai_token-act-pair · gpt-3.5-turbo

phrases indicating relationships or contexts involving organizations or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ItemTracker

-0.81

oret

-0.71

po

-0.67

bot

-0.67

WARN

-0.66

PsyNetMessage

-0.66

proof

-0.65

lopp

-0.65

blind

-0.65

wav

-0.65

POSITIVE LOGITS

 respectively

1.66

 alike

1.42

 whichever

0.98

 respective

0.85

depending

0.82

 various

0.76

 varied

0.72

 diverse

0.71

 both

0.68

 each

0.64

Activations Density 0.681%

phrases indicating instructions or guidelines

phrases indicating relationships or contexts involving organizations or systems

No Comments

No Known Activations