INDEX

Explanations

phrases with verbs associated with communication, decision-making, and actions

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

oris

-0.69

jah

-0.66

ãĤ«

-0.66

Tes

-0.63

Screenshot

-0.63

ãĥ©ãĥ³

-0.62

ika

-0.61

ja

-0.61

ãĥ£

-0.60

Crystal

-0.58

POSITIVE LOGITS

 anyways

1.24

 anyway

1.24

 AFTER

1.12

 uncond

1.07

 ourselves

1.07

 yourself

1.05

 voluntarily

1.00

 sooner

0.98

 outright

0.98

 herself

0.98

Activations Density 0.592%

phrases with verbs associated with communication, decision-making, and actions

No Comments

No Known Activations