INDEX

Explanations

phrases that express uncertainty or contemplation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 destro

-0.61

ò

-0.57

 livest

-0.57

 helicop

-0.57

ñ

-0.55

 volunte

-0.55

 contracted

-0.54

 subur

-0.53

 challeng

-0.52

 frontline

-0.51

POSITIVE LOGITS

why

0.64

 JUSTICE

0.63

 tonight

0.61

 conclusions

0.61

WHY

0.60

 understatement

0.59

 conclusion

0.59

 yourselves

0.58

 HERE

0.58

 here

0.58

Activations Density 0.623%

phrases that express uncertainty or contemplation

No Comments

No Known Activations