INDEX

Explanations

phrases related to rehabilitation and treatment

oai_token-act-pair · gpt-3.5-turbo

terms related to rehabilitation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Dad

-0.72

 minute

-0.70

 Flan

-0.68

Parent

-0.67

 Gerry

-0.62

 Mysterious

-0.61

++++++++++++++++

-0.61

meal

-0.61

xt

-0.60

 Irish

-0.60

POSITIVE LOGITS

ilitation

1.58

ilitating

1.45

 rehabilitation

1.22

 Rehab

1.14

 rehab

1.13

 rehabilit

1.08

 therap

0.93

spective

0.91

apist

0.89

eatures

0.89

Activations Density 0.010%

phrases related to rehabilitation and treatment

terms related to rehabilitation

No Comments

No Known Activations