INDEX

Explanations

terms related to beneficial or detrimental impacts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

buck

-0.80

illon

-0.75

oute

-0.67

ariat

-0.67

ebus

-0.64

hyde

-0.63

obook

-0.61

ÅŁ

-0.61

raped

-0.61

 Sina

-0.58

POSITIVE LOGITS

 effects

0.89

 synerg

0.88

effects

0.84

 outcomes

0.78

 consequences

0.78

 outweigh

0.77

 consequence

0.76

 aspects

0.73

 influence

0.73

icial

0.72

Activations Density 0.052%

terms related to beneficial or detrimental impacts

No Comments

No Known Activations