INDEX

Explanations

mentions of the word "Megan"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "gan."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ensions

-0.68

 prejudice

-0.64

ension

-0.63

 differential

-0.62

acter

-0.62

ractive

-0.62

 losers

-0.59

 recourse

-0.58

 periodic

-0.57

topic

-0.56

POSITIVE LOGITS

igans

1.13

cia

1.11

za

1.09

igan

1.08

ese

1.04

non

1.01

ogly

1.00

zie

1.00

agher

0.98

elle

0.97

Activations Density 0.064%

mentions of the word "Megan"

occurrences of the term "gan."

No Comments

No Known Activations