INDEX

Explanations

references to locations or nationalities

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bound

-0.61

ez

-0.61

ALT

-0.61

birth

-0.59

ept

-0.59

drawn

-0.57

blood

-0.57

ceptive

-0.57

 compassionate

-0.56

States

-0.55

POSITIVE LOGITS

oglu

0.88

theless

0.82

owsky

0.79

;;;;;;;;;;;;

0.77

ueller

0.77

aja

0.77

sov

0.76

ulhu

0.75

nikov

0.75

oulos

0.75

Activations Density 0.065%

references to locations or nationalities

No Comments

No Known Activations