INDEX

Explanations

phrases introducing a new piece of information or concept

oai_token-act-pair · gpt-3.5-turbo

the indefinite articles "a" and "an" 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

alties

-0.86

['

-0.80

agents

-0.77

anism

-0.71

aos

-0.71

doms

-0.71

attacks

-0.69

 occupants

-0.68

 Izan

-0.66

bars

-0.66

POSITIVE LOGITS

 snippet

1.24

 glimpse

1.22

 rundown

1.18

 snapshot

1.08

 quick

1.06

few

1.04

 couple

1.03

 handy

1.02

 screenshot

1.02

 reminder

1.02

Activations Density 0.166%

phrases introducing a new piece of information or concept

the indefinite articles "a" and "an" in various contexts

No Comments

No Known Activations