INDEX

Explanations

mentions of the name "Mos" with decreasing levels of importance indicated by the activation values

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Mos"

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istics

-0.73

 Thumbnails

-0.68

ordinate

-0.66

 Archdemon

-0.65

 Rite

-0.64

OWN

-0.62

istically

-0.62

 Volks

-0.61

ocument

-0.61

istical

-0.60

POSITIVE LOGITS

aic

1.21

quit

1.17

bian

1.05

qu

1.04

ques

1.01

lems

1.00

daq

0.99

adiq

0.99

cos

0.99

kov

0.98

Activations Density 0.027%

mentions of the name "Mos" with decreasing levels of importance indicated by the activation values

mentions of the name "Mos"

No Comments

No Known Activations