INDEX

Explanations

words related to tops or top layers

oai_token-act-pair · gpt-3.5-turbo

references to 'top' 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 warr

-0.71

[|

-0.66

 irregular

-0.61

arij

-0.60

riages

-0.60

 consent

-0.59

 Arri

-0.59

 discern

-0.59

 Gaul

-0.58

 Catal

-0.57

POSITIVE LOGITS

top

1.17

most

1.08

tops

1.05

mast

1.01

ICS

0.99

ographical

0.97

TOP

0.96

bottom

0.96

ography

0.92

Top

0.90

Activations Density 0.015%

words related to tops or top layers

references to 'top' in various contexts

No Comments

No Known Activations