INDEX

Explanations

references to specific guidelines or sets of rules

oai_token-act-pair · gpt-3.5-turbo

references to guidelines or rules

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

 spect

-0.72

some

-0.68

parts

-0.67

joy

-0.66

asts

-0.65

skill

-0.63

 Roses

-0.63

vict

-0.62

jiang

-0.61

POSITIVE LOGITS

 guidelines

1.11

 guideline

0.98

idelines

0.97

 Guidelines

0.89

 governing

0.89

DragonMagazine

0.86

 dictates

0.79

 enforced

0.78

 regulating

0.77

 dictate

0.76

Activations Density 0.029%

references to specific guidelines or sets of rules

references to guidelines or rules

No Comments

No Known Activations