INDEX

Explanations

phrases with the word "draw" followed by a concept or action

oai_token-act-pair · gpt-3.5-turbo

phrases that mention drawing a distinction or a line, particularly in a metaphorical or critical sense

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

nown

-0.81

uploads

-0.75

ESA

-0.69

ailability

-0.66

Ü

-0.65

 unden

-0.64

 Atkins

-0.63

ãĥĺ

-0.63

 Inher

-0.62

atever

-0.61

POSITIVE LOGITS

 conclusion

0.79

ire

0.77

 conclusions

0.76

 attention

0.71

ogly

0.69

 straw

0.68

circle

0.68

umbn

0.67

brid

0.67

 Dull

0.66

Activations Density 0.114%

phrases with the word "draw" followed by a concept or action

phrases that mention drawing a distinction or a line, particularly in a metaphorical or critical sense

No Comments

No Known Activations