INDEX

Explanations

words related to medical conditions and medications, specifically focusing on allergies and pharmaceuticals

oai_token-act-pair · gpt-3.5-turbo

terms related to allergies and their associated conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

hower

-0.82

 Citation

-0.73

EStream

-0.64

²¾

-0.63

earcher

-0.61

ĸļ

-0.60

 rapist

-0.60

 Paran

-0.57

 Sergeant

-0.57

Battery

-0.57

POSITIVE LOGITS

atives

0.79

 except

0.77

theless

0.76

imensional

0.69

nance

0.69

ergic

0.69

together

0.68

 whatsoever

0.66

iance

0.65

iances

0.64

Activations Density 0.069%

words related to medical conditions and medications, specifically focusing on allergies and pharmaceuticals

terms related to allergies and their associated conditions

No Comments

No Known Activations