INDEX

Explanations

instances where the word "you" is followed by particular sentiments or commands

oai_token-act-pair · gpt-3.5-turbo

second-person 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

¿½

-0.73

Attach

-0.64

Materials

-0.63

ãĤ´ãĥ³

-0.63

assembly

-0.61

OTOS

-0.55

inders

-0.54

 Inside

-0.54

 Promotion

-0.52

amaz

-0.52

POSITIVE LOGITS

're

1.39

'll

1.33

've

1.23

owe

1.23

 SHOULD

1.18

 ought

1.17

 deserve

1.17

'd

1.14

 shouldn

1.12

 should

1.08

Activations Density 0.140%

instances where the word "you" is followed by particular sentiments or commands

second-person references

No Comments

No Known Activations