INDEX

Explanations

words associated with medical treatment or care

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "treat."

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

wick

-0.75

âĢİ

-0.72

 Voices

-0.64

 Specifications

-0.64

ews

-0.59

lights

-0.59

media

-0.59

bro

-0.58

leigh

-0.57

builder

-0.57

POSITIVE LOGITS

 treat

3.91

 treats

2.39

 Treat

2.38

 treating

2.04

 treated

1.88

 treatment

1.57

treated

1.32

 Treatment

1.30

 treatments

1.26

treatment

1.19

Activations Density 0.006%

words associated with medical treatment or care

occurrences of the word "treat."

No Comments

No Known Activations