INDEX

Explanations

phrases related to completeness or abundance

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "whole" 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 @ 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

marks

-0.79

rations

-0.71

effects

-0.67

tips

-0.66

abilities

-0.66

INO

-0.64

evidence

-0.64

eus

-0.62

placed

-0.62

inates

-0.61

POSITIVE LOGITS

 bunch

1.27

 slew

1.24

heartedly

1.22

lot

1.17

hearted

1.05

 raft

1.05

 heap

1.02

 swath

1.01

 host

0.92

new

0.87

Activations Density 0.038%

phrases related to completeness or abundance

references to the concept of "whole" in various contexts

No Comments

No Known Activations