INDEX

Explanations

phrases related to technology terms and agreements

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

Data Type

torch.float32

Hook Point

blocks.8.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

ento

-0.72

 baseman

-0.69

 bombard

-0.66

 horr

-0.64

 tempted

-0.64

 reassure

-0.63

earances

-0.61

 horrified

-0.61

reau

-0.61

 reassured

-0.61

POSITIVE LOGITS

1.02

1.01

"#

0.99

``

0.91

0.90

''

0.80

"'

0.77

\"

0.77

Î

0.77

"_

0.75

Activations Density 4.762%

phrases related to technology terms and agreements

No Comments

No Known Activations