INDEX

Explanations

instances where something is mentioned as happening later in a specific context

oai_token-act-pair · gpt-3.5-turbo

the phrase "in the" followed by numbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

-0.74

afety

-0.68

OTH

-0.64

 inherit

-0.64

bryce

-0.61

sworth

-0.61

CLASSIFIED

-0.60

username

-0.60

GROUND

-0.59

llor

-0.58

POSITIVE LOGITS

humane

1.19

offensive

1.15

ordinate

1.07

clusions

1.06

clus

1.06

 lieu

1.04

effic

1.04

iqu

1.03

roads

1.03

organic

1.03

Activations Density 0.153%

instances where something is mentioned as happening later in a specific context

the phrase "in the" followed by numbers

No Comments

No Known Activations