INDEX

Explanations

phrases related to causality or explanations

oai_token-act-pair · gpt-3.5-turbo

causal relationships or conditions leading to specific outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

asp

-0.72

=-=-=-=-=-=-=-=-

-0.71

arro

-0.69

oola

-0.68

adel

-0.68

atari

-0.68

ille

-0.67

 Sheep

-0.67

anon

-0.66

arest

-0.65

POSITIVE LOGITS

 diligence

1.13

giving

0.81

lling

0.74

 dilig

0.71

itiz

0.69

 inhibition

0.67

cling

0.66

llers

0.66

 solely

0.65

 cancell

0.65

Activations Density 0.027%

phrases related to causality or explanations

causal relationships or conditions leading to specific outcomes

No Comments

No Known Activations