INDEX

Explanations

medical terms related to imaging techniques and pathological conditions

oai_token-act-pair · gpt-3.5-turbo

terms related to biological and medical research techniques and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

guard

-0.68

lain

-0.68

inness

-0.66

esson

-0.66

 allegiance

-0.65

inoa

-0.65

FUL

-0.65

esses

-0.65

enhagen

-0.64

ingham

-0.64

POSITIVE LOGITS

ascus

0.81

ibaba

0.80

nec

0.80

spect

0.75

ctive

0.74

 antigen

0.72

ures

0.71

cular

0.68

tical

0.67

apy

0.66

Activations Density 0.067%

medical terms related to imaging techniques and pathological conditions

terms related to biological and medical research techniques and conditions

No Comments

No Known Activations