INDEX

Explanations

phrases related to legal matters and social justice issues

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

 promise

-0.70

 Wick

-0.66

might

-0.63

 wander

-0.63

 Might

-0.62

 traverse

-0.61

 mark

-0.60

 roam

-0.59

 dare

-0.59

 claim

-0.59

POSITIVE LOGITS

 able

1.26

 judged

1.10

 regarded

1.09

 considered

1.08

fitting

1.08

 viewed

1.04

 treated

1.04

leeve

1.01

 construed

0.99

 replaced

0.97

Activations Density 0.604%

phrases related to legal matters and social justice issues

No Comments

No Known Activations