INDEX

Explanations

words related to medical positions and titles, such as "MD" and "surgeons"

oai_token-act-pair · gpt-3.5-turbo

the abbreviation "MD" in various contexts, likely indicating medical doctors or degrees

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

stall

-0.85

acters

-0.82

flies

-0.76

neau

-0.75

nel

-0.72

nces

-0.72

angered

-0.72

asonable

-0.72

ãĥ¼ãĥ³

-0.71

berus

-0.71

POSITIVE LOGITS

iamond

0.94

ouble

0.83

aughter

0.82

iate

0.81

MD

0.73

ayer

0.72

ORPG

0.71

eeds

0.70

ART

0.69

mology

0.68

Activations Density 0.014%

words related to medical positions and titles, such as "MD" and "surgeons"

the abbreviation "MD" in various contexts, likely indicating medical doctors or degrees

No Comments

No Known Activations