INDEX

Explanations

information related to government agencies and organizations

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

olars

-0.75

uncture

-0.74

thora

-0.72

tics

-0.68

acca

-0.68

Sov

-0.68

 suffice

-0.68

erity

-0.67

chio

-0.67

zees

-0.67

POSITIVE LOGITS

 acronym

0.95

quartered

0.94

 powerhouse

0.89

 successor

0.88

 headquartered

0.85

 famed

0.84

 abbre

0.83

 renowned

0.82

 formerly

0.82

formerly

0.82

Activations Density 1.666%

information related to government agencies and organizations

No Comments

No Known Activations