INDEX

Explanations

direct speech within text, particularly quotes

oai_token-act-pair · gpt-3.5-turbo

references to information or explanations provided by others

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

models

-0.81

±

-0.77

§

-0.71

¢

-0.71

¥µ

-0.70

µ

-0.70

´

-0.66

Downloadha

-0.66

cible

-0.65

conn

-0.63

POSITIVE LOGITS

:-

0.91

 nutshell

0.90

 excerpt

0.73

:#

0.72

 illustrating

0.71

 glim

0.70

:(

0.70

*:

0.69

 illustrate

0.68

:[

0.66

Activations Density 0.373%

direct speech within text, particularly quotes

references to information or explanations provided by others

No Comments

No Known Activations