INDEX

Explanations

negative sentiments and emotions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

esson

-0.71

èĢħ

-0.69

 guiIcon

-0.69

 Unicode

-0.69

 initials

-0.68

ABE

-0.68

 Congressional

-0.67

 Schumer

-0.66

 identifiers

-0.65

POSITIVE LOGITS

condition

1.19

filled

1.19

fuel

1.15

floor

1.15

based

1.14

bath

1.13

water

1.13

tight

1.13

storm

1.12

tun

1.11

Activations Density 0.031%

negative sentiments and emotions

No Comments

No Known Activations