INDEX

Explanations

references to medical prescriptions or instructions

oai_token-act-pair · gpt-3.5-turbo

words related to the act of prescribing medications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Squid

-0.79

 squid

-0.78

Grab

-0.72

cade

-0.69

Buzz

-0.67

ager

-0.66

tery

-0.66

Ink

-0.65

TG

-0.65

aunch

-0.64

POSITIVE LOGITS

 prescribed

3.44

 prescribe

2.86

 prescribing

2.46

 prescriptions

2.13

 presc

1.95

cribed

1.58

 prescription

1.54

 administered

1.13

 promulg

1.09

 recommended

1.07

Activations Density 0.010%

references to medical prescriptions or instructions

words related to the act of prescribing medications

No Comments

No Known Activations