INDEX

Explanations

warnings or alerts

oai_token-act-pair · gpt-3.5-turbo

mentions of warnings

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

hedral

-0.82

morph

-0.78

animate

-0.77

growth

-0.70

ablished

-0.68

ophon

-0.67

rencies

-0.66

rafted

-0.66

anova

-0.65

artney

-0.65

POSITIVE LOGITS

warning

1.00

 warning

0.96

 warnings

0.90

 Warn

0.89

 Warning

0.87

 disclaimer

0.85

 Signs

0.81

 warns

0.80

warn

0.79

 warn

0.76

Activations Density 0.030%

warnings or alerts

mentions of warnings

No Comments

No Known Activations