INDEX

Explanations

terms related to reasoning, certainty, and drawing conclusions based on evidence and analysis

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ophon

-0.58

nas

-0.55

 talents

-0.53

cler

-0.49

bid

-0.49

amen

-0.49

ansk

-0.47

inventory

-0.47

 surn

-0.47

 Leban

-0.47

POSITIVE LOGITS

0.68

 conclusion

0.67

 conclusions

0.62

éĸ

0.55

lessly

0.54

iveness

0.52

 regarding

0.52

ä½ľ

0.51

 Conclusion

0.50

 drawn

0.49

Activations Density 5.069%

terms related to reasoning, certainty, and drawing conclusions based on evidence and analysis

No Comments

No Known Activations