INDEX

Explanations

adjectives with positive or negative sentiment

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

çīĪ

-0.75

ħĭ

-0.70

undai

-0.70

ADRA

-0.70

ORTS

-0.69

aucuses

-0.67

Xi

-0.67

rosso

-0.66

İĭ

-0.66

OULD

-0.66

POSITIVE LOGITS

 everywhere

0.95

 nowadays

0.92

 inherently

0.78

 intrinsically

0.78

 predictor

0.76

 unto

0.76

 commodity

0.75

 motiv

0.70

↵Âł

0.70

 mistress

0.69

Activations Density 0.291%

adjectives with positive or negative sentiment

No Comments

No Known Activations