INDEX

Explanations

phrases or terms related to speech or verbal communication

oai_token-act-pair · gpt-3.5-turbo

the word "ep" and its variations, suggesting a focus on specific references or notations commonly used in transcripts or quoted speech

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 nomine

-0.75

Rid

-0.72

Frameworks

-0.70

 AMERICA

-0.66

 Amend

-0.66

 referen

-0.66

Ô

-0.66

é¾

-0.64

ãĤ´ãĥ³

-0.62

 Razer

-0.62

POSITIVE LOGITS

isodes

1.29

hemer

1.16

iphany

1.10

rint

1.05

odcast

1.03

artment

1.02

osit

0.93

eta

0.93

ublic

0.92

och

0.91

Activations Density 0.008%

phrases or terms related to speech or verbal communication

the word "ep" and its variations, suggesting a focus on specific references or notations commonly used in transcripts or quoted speech

No Comments

No Known Activations