INDEX

Explanations

phrases related to political and geopolitical discussions

oai_token-act-pair · gpt-3.5-turbo

symbols or special characters that indicate emphasis or a specific style of presentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 cyan

-0.77

 decomp

-0.76

 floating

-0.73

 dirt

-0.73

 shrouded

-0.73

 fertil

-0.72

 spinning

-0.71

 scattering

-0.70

 shack

-0.70

 minim

-0.69

POSITIVE LOGITS

º

0.96

£

0.94

ĸļ

0.84

¹

0.83

Serv

0.83

 Ibid

0.79

catentry

0.79

Ī

0.79

Sch

0.77

âĢķ

0.77

Activations Density 0.237%

phrases related to political and geopolitical discussions

symbols or special characters that indicate emphasis or a specific style of presentation

No Comments

No Known Activations