INDEX

Explanations

names or terms related to a specific entity "Tan"

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ãĤ¡

-0.70

 Verd

-0.66

Ö¼

-0.65

 Madison

-0.65

 laure

-0.64

âĸĪâĸĪ

-0.64

Madison

-0.64

pton

-0.62

 Mellon

-0.62

sis

-0.62

POSITIVE LOGITS

nery

1.33

ning

1.18

jong

1.01

oco

1.00

geon

0.99

uki

0.97

quer

0.97

zan

0.96

jug

0.92

agra

0.88

Activations Density 0.050%

names or terms related to a specific entity "Tan"

No Comments

No Known Activations