INDEX

Explanations

words related to bases or foundational elements

oai_token-act-pair · gpt-3.5-turbo

references to a "base" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

NESS

-0.84

oso

-0.77

estern

-0.76

apes

-0.73

icut

-0.72

icer

-0.69

asar

-0.67

istor

-0.67

istors

-0.67

inness

-0.67

POSITIVE LOGITS

camp

1.16

plate

0.91

plates

0.91

ball

0.89

BALL

0.78

 erosion

0.77

band

0.75

plan

0.73

Layer

0.72

bridge

0.71

Activations Density 0.017%

words related to bases or foundational elements

references to a "base" in various contexts

No Comments

No Known Activations