INDEX

Explanations

verbs or phrases related to actions or activities

oai_token-act-pair · gpt-3.5-turbo

verbs associated with understanding or communication

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

 vulner

-0.69

emale

-0.68

 contrace

-0.66

ilaterally

-0.66

 advoc

-0.66

stood

-0.66

 helicop

-0.64

 corrid

-0.62

 Vaugh

-0.62

atever

-0.61

POSITIVE LOGITS

Psy

0.63

actionDate

0.62

 Vulkan

0.62

CRIPTION

0.60

CLUD

0.57

USB

0.54

lihood

0.54

 False

0.54

 Coverage

0.54

Ple

0.53

Activations Density 0.139%

verbs or phrases related to actions or activities

verbs associated with understanding or communication

No Comments

No Known Activations