INDEX

Explanations

personal commands or instructions

oai_token-act-pair · gpt-3.5-turbo

the second-person pronoun "you" and its variants

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

ãĥ³ãĤ¸

-0.68

 Dimension

-0.65

 Submission

-0.63

 Ambro

-0.62

burgh

-0.60

interstitial

-0.58

 Gamb

-0.58

ilib

-0.58

ogenesis

-0.57

 Rosenthal

-0.57

POSITIVE LOGITS

're

1.28

 wanna

1.01

 want

0.99

've

0.97

 guessed

0.94

 guys

0.90

 wish

0.87

 subscribed

0.86

 weren

0.84

 subscribe

0.84

Activations Density 0.094%

personal commands or instructions

the second-person pronoun "you" and its variants

No Comments

No Known Activations