INDEX

Explanations

names of people and locations

oai_token-act-pair · gpt-3.5-turbo

specific structured data or lists, particularly those involving names, numbers, and statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

conservancy

-0.69

ãĥ¼ãĤ¯

-0.66

 Eater

-0.61

ensional

-0.58

 WARN

-0.56

terness

-0.54

 counterpart

-0.54

:/

-0.53

uggest

-0.51

 predecessor

-0.50

POSITIVE LOGITS

etc

0.90

 respectively

0.80

etc

0.74

 Lastly

0.66

Jr

0.63

};

0.62

Conclusion

0.60

âĢ¦)

0.59

oral

0.59

inson

0.59

Activations Density 0.704%

names of people and locations

specific structured data or lists, particularly those involving names, numbers, and statistics

No Comments

No Known Activations