INDEX

Explanations

phrases related to legal decisions or actions.

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

icons

-0.90

uded

-0.71

 sophistication

-0.67

 passers

-0.66

idable

-0.66

tips

-0.66

bits

-0.65

acons

-0.64

ortun

-0.64

icious

-0.63

POSITIVE LOGITS

 postpone

1.17

 adjourn

1.05

 disband

1.03

 postp

1.02

 defer

0.98

 relocate

0.96

 proceed

0.95

 withdraw

0.95

 pursue

0.87

 resign

0.85

Activations Density 2.059%

**phrases related to legal decisions or actions.**

No Comments

No Known Activations

phrases related to legal decisions or actions.