INDEX

Explanations

phrases related to authoritative statements or reports

oai_token-act-pair · gpt-3.5-turbo

pronouns referring to individuals, particularly in a possessive context

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

 INCLUD

-0.72

otin

-0.70

import

-0.68

=\"

-0.66

Æ

-0.66

(?,

-0.64

 Researchers

-0.63

 Iranians

-0.62

vantage

-0.61

hn

-0.61

POSITIVE LOGITS

own

1.23

 autobiography

1.14

 keynote

1.08

 introductory

1.05

 memoir

1.04

 remarks

0.96

 maiden

0.92

 commentary

0.90

Own

0.90

 speech

0.90

Activations Density 0.103%

phrases related to authoritative statements or reports

pronouns referring to individuals, particularly in a possessive context

No Comments

No Known Activations