INDEX

Explanations

locations

oai_token-act-pair · gpt-3.5-turbo

references to locations and institutions within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ousands

-0.64

luster

-0.62

mble

-0.58

ancies

-0.58

leys

-0.55

nces

-0.55

}}

-0.55

ults

-0.54

ischer

-0.52

OHN

-0.52

POSITIVE LOGITS

 that

1.40

that

1.35

 whereby

1.25

 wherein

1.21

 which

1.13

 called

1.08

 whose

1.06

 titled

1.05

 THAT

1.04

 where

1.02

Activations Density 1.053%

locations

references to locations and institutions within various contexts

No Comments

No Known Activations

locations

references to locations and institutions within various contexts

No Comments

No Known Activations