INDEX

Explanations

terms related to fire or heat

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

ties

-1.01

odox

-0.85

ager

-0.81

alian

-0.80

ifies

-0.75

aged

-0.73

itism

-0.73

 Leilan

-0.73

psons

-0.73

chery

-0.73

POSITIVE LOGITS

 Sands

0.84

 sands

0.81

steel

0.81

 fists

0.77

 Skies

0.75

 flame

0.75

erno

0.74

 fury

0.74

hot

0.74

 metal

0.73

Activations Density 0.021%

terms related to fire or heat

No Comments

No Known Activations