INDEX

Explanations

parentheses

oai_token-act-pair · gpt-3.5-turbo

parenthetical expressions or references

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

 nutrient

-0.75

 semblance

-0.75

 routine

-0.74

 antiv

-0.73

 appropri

-0.72

 absor

-0.70

 pill

-0.70

 transcend

-0.70

 wallpaper

-0.68

 pine

-0.68

POSITIVE LOGITS

âĢ¦)

1.45

laughs

1.39

...)

1.29

Laughs

1.29

sic

1.25

emphasis

1.23

hide

1.15

laughter

1.10

?)

1.10

++)

1.09

Activations Density 0.095%

parentheses

parenthetical expressions or references

No Comments

No Known Activations