INDEX

Explanations

incorrect or inaccurate information

oai_token-act-pair · gpt-3.5-turbo

words related to wrongful accusations and misdiagnoses

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

iments

-0.85

ĸļ

-0.77

iment

-0.75

OTOS

-0.75

iens

-0.70

ista

-0.70

orem

-0.67

istas

-0.66

 horizont

-0.64

 Extensions

-0.64

POSITIVE LOGITS

 mistaken

0.88

 accuse

0.85

 inaccurate

0.83

 priced

0.81

ãĤ©

0.80

 inflated

0.78

 imprisoned

0.75

 accused

0.75

 imprison

0.75

 guiActiveUn

0.75

Activations Density 0.026%

incorrect or inaccurate information

words related to wrongful accusations and misdiagnoses

No Comments

No Known Activations