INDEX

Explanations

occurrences of commas in lists or descriptions

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

sburg

-0.81

pec

-0.80

lyn

-0.74

olves

-0.70

agin

-0.69

yon

-0.69

veland

-0.68

yan

-0.67

½

-0.67

urus

-0.66

POSITIVE LOGITS

 which

1.12

 dubbed

0.97

 although

0.93

 albeit

0.93

 namely

0.88

which

0.88

 noting

0.84

 though

0.83

 citing

0.82

 nicknamed

0.82

Activations Density 0.217%

occurrences of commas in lists or descriptions

No Comments

No Known Activations