INDEX

Explanations

occurrences of the letter 'O' with relatively high activations

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

theless

-0.70

sburgh

-0.64

inois

-0.63

iard

-0.61

 textures

-0.61

awed

-0.61

 subp

-0.61

 identifiable

-0.60

 hitherto

-0.60

 condem

-0.60

POSITIVE LOGITS

zzi

1.27

lymp

1.16

OPS

1.15

JO

1.12

LET

1.08

XY

1.08

VE

1.06

ISE

1.06

OP

1.05

LL

1.04

Activations Density 0.028%

occurrences of the letter 'O' with relatively high activations

No Comments

No Known Activations