INDEX

Explanations

descriptions of things hidden or located beneath something

oai_token-act-pair · gpt-3.5-turbo

references to things that are concealed or hidden below the surface

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

eln

-0.76

iji

-0.70

itar

-0.70

ordan

-0.69

yah

-0.69

ebus

-0.69

zai

-0.68

za

-0.67

MM

-0.67

eller

-0.65

POSITIVE LOGITS

neath

1.00

pins

0.90

eatures

0.86

sea

0.84

İĭ

0.84

 layers

0.83

ĸļ

0.83

lip

0.76

ĨĴ

0.75

¥ŀ

0.74

Activations Density 0.014%

descriptions of things hidden or located beneath something

references to things that are concealed or hidden below the surface

No Comments

No Known Activations