INDEX

Explanations

phrases related to change or redirection of actions or focus

oai_token-act-pair · gpt-3.5-turbo

the conjunction "and" 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 @ 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

BAT

-0.73

aw

-0.71

ACP

-0.69

ACT

-0.64

 Patron

-0.64

aws

-0.63

ULTS

-0.63

Mich

-0.63

Penn

-0.62

Viz

-0.62

POSITIVE LOGITS

 preferring

1.09

 opting

1.04

 replaced

1.03

 instead

1.01

 simplify

1.00

 concentrate

0.99

 replaces

0.96

 Instead

0.93

 prefers

0.92

 embrace

0.92

Activations Density 0.372%

phrases related to change or redirection of actions or focus

the conjunction "and" in various contexts

No Comments

No Known Activations