INDEX

Explanations

instances where the word "in" is emphasized in the context

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in" that signify contextual references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

llor

-0.80

rina

-0.72

linger

-0.72

quad

-0.66

ŀ

-0.65

angered

-0.64

LEASE

-0.64

edin

-0.64

cha

-0.63

NOW

-0.63

POSITIVE LOGITS

effic

1.55

efficiency

1.29

 relation

1.29

 regards

1.27

clusions

1.21

 terms

1.15

 lieu

1.15

 regard

1.05

escap

0.99

 favour

0.99

Activations Density 0.268%

instances where the word "in" is emphasized in the context

occurrences of the word "in" that signify contextual references

No Comments

No Known Activations