INDEX

Explanations

mentions of "live-action" or related terms in texts

oai_token-act-pair · gpt-3.5-turbo

terms related to live-action content and streaming events

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

Ö¼

-0.87

=-=-

-0.70

jee

-0.68

²¾

-0.67

 actionGroup

-0.65

etsy

-0.65

ãĥ¼ãĥĨ

-0.63

ãĤ©

-0.63

jit

-0.61

utsu

-0.61

POSITIVE LOGITS

jriwal

0.74

 spectator

0.71

 Marse

0.64

Unity

0.64

gypt

0.64

 breathing

0.62

opter

0.62

 spectators

0.61

 streamed

0.61

naissance

0.61

Activations Density 0.118%

mentions of "live-action" or related terms in texts

terms related to live-action content and streaming events

No Comments

No Known Activations