INDEX

Explanations

phrases related to personal experiences and reflections

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

.","

-0.76

 ÂŃ

-0.75

 jointly

-0.74

 concess

-0.73

 lawfully

-0.70

 thereto

-0.69

 multim

-0.68

 practicable

-0.68

 locally

-0.68

 biod

-0.67

POSITIVE LOGITS

fuck

0.92

laughs

0.90

Anyway

0.90

yrics

0.90

yeah

0.87

 Shit

0.85

eeee

0.85

Laughs

0.84

 fuck

0.84

laughter

0.84

Activations Density 0.824%

phrases related to personal experiences and reflections

No Comments

No Known Activations