INDEX

Explanations

phrases related to political and governmental actions

oai_token-act-pair · gpt-3.5-turbo

topics related to political actions and their consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 alot

-0.77

 doesnt

-0.71

 brill

-0.70

 negro

-0.69

 english

-0.69

LOT

-0.68

 didnt

-0.68

 ALWAYS

-0.66

 fucking

-0.65

 ONLY

-0.65

POSITIVE LOGITS

0.85

 amid

0.84

 âĢķ

0.84

 nationwide

0.81

—

0.80

 politically

0.80

 federally

0.79

.—

0.79

 nationally

0.77

—

0.77

Activations Density 0.546%

phrases related to political and governmental actions

topics related to political actions and their consequences

No Comments

No Known Activations