INDEX

Explanations

words related to specific names or brands that include 'ye' or 'yer'

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_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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ITY

-0.89

ORY

-0.81

ctica

-0.81

ITIES

-0.80

bons

-0.78

lished

-0.76

INESS

-0.76

ENA

-0.74

matic

-0.70

ahime

-0.69

POSITIVE LOGITS

terday

0.94

eder

0.83

ldon

0.79

arthed

0.78

asure

0.77

lde

0.76

xt

0.75

0.74

asing

0.70

lda

0.70

Activations Density 0.016%

words related to specific names or brands that include 'ye' or 'yer'

No Comments

No Known Activations