INDEX

Explanations

phrases related to boundaries or limits being crossed

oai_token-act-pair · gpt-3.5-turbo

references to limits or boundaries

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

CVE

-0.85

cia

-0.84

sylvania

-0.77

icably

-0.74

igl

-0.74

illac

-0.67

ilton

-0.67

olute

-0.67

reditary

-0.64

tis

-0.64

POSITIVE LOGITS

aments

1.13

backer

1.11

 separating

0.99

 dividing

0.96

ups

0.91

breaking

0.91

ages

0.83

 crossings

0.82

breaks

0.82

 connecting

0.81

Activations Density 0.040%

phrases related to boundaries or limits being crossed

references to limits or boundaries

No Comments

No Known Activations