INDEX

Explanations

medical conditions and terms related to a variety of diseases

oai_token-act-pair · gpt-3.5-turbo

phrases that list or describe multiple items or examples in a text

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

scribe

-0.69

arde

-0.68

Į

-0.67

atre

-0.67

gged

-0.65

ady

-0.61

 stricken

-0.61

ahime

-0.60

 itself

-0.60

emaker

-0.60

POSITIVE LOGITS

 including

1.36

 Including

1.30

including

1.19

 besides

1.02

 namely

0.95

 ranging

0.95

 culminating

0.90

 includ

0.88

 except

0.87

except

0.84

Activations Density 0.781%

medical conditions and terms related to a variety of diseases

phrases that list or describe multiple items or examples in a text

No Comments

No Known Activations