INDEX

Explanations

accusations or allegations

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

emetery

-0.93

Interstitial

-0.77

ONSORED

-0.76

ahime

-0.74

nton

-0.74

chedel

-0.69

mid

-0.69

oppy

-0.68

ARCH

-0.67

accompan

-0.67

POSITIVE LOGITS

 accusing

0.92

 accuse

0.82

him

0.81

 abusers

0.81

ologists

0.80

 leve

0.73

 blaming

0.73

 rapists

0.73

 accus

0.72

onia

0.71

Activations Density 0.033%

accusations or allegations

No Comments

No Known Activations