INDEX

Explanations

names or terms related to specific individuals or entities

oai_token-act-pair · gpt-3.5-turbo

specific names or terms related to cultural or geographical references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

deck

-0.75

trace

-0.68

mable

-0.66

sheet

-0.64

said

-0.62

rified

-0.62

matically

-0.61

 fairness

-0.61

Forward

-0.61

driving

-0.61

POSITIVE LOGITS

ÅŁ

1.38

ppa

1.33

á¹

1.25

qa

1.22

qs

1.18

Ã±

1.16

ÄŁ

1.15

qi

1.14

pta

1.12

1.10

Activations Density 0.180%

names or terms related to specific individuals or entities

specific names or terms related to cultural or geographical references

No Comments

No Known Activations