INDEX

Explanations

terms related to secretive operations or activities

oai_token-act-pair · gpt-3.5-turbo

references to secrecy and covert operations

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

versely

-0.77

ardless

-0.72

itant

-0.72

reau

-0.70

hetically

-0.68

cially

-0.66

negie

-0.64

uliffe

-0.64

 spons

-0.64

Interstitial

-0.63

POSITIVE LOGITS

ãĤ»

0.69

clave

0.67

glers

0.66

rencies

0.63

ITNESS

0.63

 lurking

0.62

 secret

0.61

geries

0.61

iuses

0.61

 Grimoire

0.60

Activations Density 0.398%

terms related to secretive operations or activities

references to secrecy and covert operations

No Comments

No Known Activations