INDEX

Explanations

words related to physical locations or persons

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ond"

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

mson

-0.85

kson

-0.65

======

-0.64

jriwal

-0.62

CLE

-0.61

GBT

-0.61

plates

-0.59

glers

-0.59

HCR

-0.57

IPS

-0.57

POSITIVE LOGITS

ragon

1.26

rive

1.02

erer

1.01

ering

0.99

orf

0.98

ition

0.96

romeda

0.93

roit

0.89

itional

0.89

imensional

0.88

Activations Density 0.028%

words related to physical locations or persons

occurrences of the substring "ond"

No Comments

No Known Activations