INDEX

Explanations

actions involving explanations or reasons

oai_token-act-pair · gpt-3.5-turbo

words associated with explaining or giving reasons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

luster

-0.81

assic

-0.69

dar

-0.67

rica

-0.67

ille

-0.66

sembly

-0.65

inates

-0.64

Pont

-0.63

itton

-0.63

oreal

-0.63

POSITIVE LOGITS

why

1.75

WHY

1.47

how

1.36

why

1.34

how

1.04

HOW

0.97

Why

0.97

Why

0.91

 what

0.90

 away

0.83

Activations Density 0.069%

actions involving explanations or reasons

words associated with explaining or giving reasons

No Comments

No Known Activations