INDEX

Explanations

quantitative rankings or positions based on performance metrics

oai_token-act-pair · gpt-3.5-turbo

terms related to ranking or assessments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ery

-0.70

 attendant

-0.67

 Leilan

-0.65

iae

-0.65

 realised

-0.63

umb

-0.62

ciation

-0.61

eway

-0.61

 Ultr

-0.58

orship

-0.56

POSITIVE LOGITS

 among

0.85

 prominently

0.84

among

0.83

âĺħâĺħ

0.82

 ninth

0.81

 favorably

0.81

 sixth

0.76

 seventh

0.75

higher

0.73

 eighth

0.72

Activations Density 0.105%

quantitative rankings or positions based on performance metrics

terms related to ranking or assessments

No Comments

No Known Activations