INDEX

Explanations

symbols and non-standard characters typically used in text encoding errors or foreign alphabets

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

 crossover

-0.63

agall

-0.62

 succession

-0.62

PSP

-0.61

 sensitivity

-0.61

 connection

-0.60

 temptation

-0.60

 draws

-0.60

 commissions

-0.59

 womb

-0.59

POSITIVE LOGITS

Ðµ

1.06

Ð¾Ð

1.04

ÑĢ

1.03

Ñĥ

1.00

Ð»

0.97

Ð¾

0.97

Ð°

0.94

ÙĨ

0.92

Ð²

0.92

¹

0.91

Activations Density 0.036%

symbols and non-standard characters typically used in text encoding errors or foreign alphabets

No Comments

No Known Activations