INDEX

Explanations

terms related to governmental approval and support

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

tein

-0.78

rete

-0.77

cientious

-0.63

urger

-0.59

ãĥ£

-0.59

gen

-0.58

 conduc

-0.57

ergy

-0.56

arius

-0.56

Kou

-0.56

POSITIVE LOGITS

SPONSORED

0.94

back

0.74

 permitting

0.72

 mandates

0.69

dogs

0.67

hyde

0.67

 backing

0.64

airs

0.63

checks

0.62

 alike

0.62

Activations Density 0.120%

terms related to governmental approval and support

No Comments

No Known Activations