INDEX

Explanations

phrases indicating collective actions or sentiments

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

 Rowe

-0.75

imum

-0.64

aryl

-0.63

¿½

-0.62

Applications

-0.61

 Citation

-0.59

ãĤ¯

-0.56

ricular

-0.56

ocl

-0.55

uates

-0.55

POSITIVE LOGITS

're

1.31

'll

1.16

've

1.13

ird

1.11

eeee

1.06

asel

1.06

bsite

1.06

 gotta

1.02

athered

0.99

IRD

0.98

Activations Density 0.143%

phrases indicating collective actions or sentiments

No Comments

No Known Activations