INDEX

Explanations

references to specific content or citations within a text

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

 wildlife

-0.79

 retard

-0.74

 flow

-0.72

 lull

-0.72

 daily

-0.72

 clin

-0.72

 intensive

-0.71

 park

-0.71

 star

-0.69

 prey

-0.69

POSITIVE LOGITS

including

1.69

although

1.66

see

1.64

which

1.64

excluding

1.61

except

1.61

albeit

1.61

usually

1.56

especially

1.54

sic

1.52

Activations Density 0.321%

references to specific content or citations within a text

No Comments

No Known Activations