INDEX

Explanations

negative sentiments or expressions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 Confederation

-0.74

 Howe

-0.66

ĨĴ

-0.64

 Borders

-0.64

 Cause

-0.63

HCR

-0.62

 existence

-0.61

 constitu

-0.60

 congr

-0.60

 Province

-0.60

POSITIVE LOGITS

down

1.17

happy

1.13

downs

1.09

out

1.07

dash

1.06

outs

1.05

starting

1.05

back

1.05

hitting

1.05

backs

1.02

Activations Density 0.035%

negative sentiments or expressions

No Comments

No Known Activations