INDEX

Explanations

terms related to data points or measurements

oai_token-act-pair · gpt-3.5-turbo

mentions of peptides or biochemical terms related to biology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

CE

-0.69

hold

-0.65

 Diablo

-0.63

 Bravo

-0.61

ISH

-0.60

hips

-0.59

 Spears

-0.58

clusive

-0.57

NRS

-0.56

Bey

-0.56

POSITIVE LOGITS

itude

1.10

icket

1.08

acular

1.08

ember

1.05

ronics

0.97

uple

0.95

eenth

0.95

itious

0.93

itudes

0.93

unia

0.93

Activations Density 0.032%

terms related to data points or measurements

mentions of peptides or biochemical terms related to biology

No Comments

No Known Activations