INDEX

Explanations

instances where someone is making a specific observation or point

oai_token-act-pair · gpt-3.5-turbo

instances of the word "noted."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

quer

-0.83

atom

-0.72

soever

-0.71

ravel

-0.71

icum

-0.68

adesh

-0.68

heed

-0.67

ascus

-0.66

WAY

-0.66

iland

-0.66

POSITIVE LOGITS

 inconsistencies

0.86

 noting

0.81

 discrepancies

0.77

how

0.75

 approving

0.74

 prominently

0.73

 similarities

0.72

unct

0.72

 favorably

0.71

 notes

0.68

Activations Density 0.042%

instances where someone is making a specific observation or point

instances of the word "noted."

No Comments

No Known Activations