INDEX

Explanations

phrases related to suffering from medical conditions

oai_token-act-pair · gpt-3.5-turbo

phrases related to suffering from health conditions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

yssey

-0.99

itures

-0.84

soDeliveryDate

-0.81

uristic

-0.80

mere

-0.79

hatt

-0.77

displayText

-0.76

asus

-0.76

iture

-0.75

leased

-0.74

POSITIVE LOGITS

 illness

1.12

 debilitating

1.11

dys

1.10

 delusions

1.07

 depression

1.07

 severe

1.06

 dementia

1.04

 acute

1.03

 anxiety

1.02

 symptoms

1.01

Activations Density 0.098%

phrases related to suffering from medical conditions

phrases related to suffering from health conditions

No Known Activations