INDEX

Explanations

references to popular culture

oai_token-act-pair · gpt-3.5-turbo

references to pop culture and its various manifestations

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

 Vera

-0.77

 humane

-0.64

heart

-0.64

Liv

-0.64

 Sheep

-0.62

Tus

-0.61

Hus

-0.61

¥

-0.61

Ric

-0.60

 Georg

-0.60

POSITIVE LOGITS

 depictions

0.83

 outlets

0.83

 trivia

0.81

 studios

0.81

culture

0.80

advertising

0.80

 references

0.80

 appropriation

0.77

 References

0.77

 satire

0.75

Activations Density 0.048%

references to popular culture

references to pop culture and its various manifestations

No Comments

No Known Activations