INDEX

Explanations

terms related to technical processes and physical phenomena

oai_token-act-pair · gpt-3.5-turbo

medical terminology related to health risks and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Advent

-0.70

 Equity

-0.68

 Memor

-0.67

 Kahn

-0.67

 Collabor

-0.67

 categ

-0.66

 understatement

-0.66

 Lithuan

-0.65

 fairness

-0.64

 Revision

-0.63

POSITIVE LOGITS

 vapor

1.23

 liqu

1.20

 oxid

1.10

 disinteg

1.07

 contamin

1.06

 explodes

1.02

 molten

1.02

 evapor

1.01

 ejected

1.01

 dissip

1.01

Activations Density 0.334%

terms related to technical processes and physical phenomena

medical terminology related to health risks and conditions

No Comments

No Known Activations