INDEX

Explanations

phrases indicating degrees of change or modification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

DragonMagazine

-0.77

aring

-0.77

IRO

-0.68

urated

-0.67

psons

-0.64

 Paste

-0.63

CRIPTION

-0.63

Ò

-0.62

rets

-0.61

ovic

-0.60

POSITIVE LOGITS

bit

0.81

 stra

0.81

heartedly

0.79

 longer

0.76

 differently

0.75

 stead

0.70

 sooner

0.69

 indist

0.68

ighter

0.67

 farther

0.67

Activations Density 0.100%

phrases indicating degrees of change or modification

No Comments

No Known Activations