INDEX

Explanations

links or commands prompting the user to view or open content

oai_token-act-pair · gpt-3.5-turbo

prompts related to viewing or accessing media content

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

ãĤ´ãĥ³

-0.86

 convol

-0.76

etheless

-0.71

writers

-0.66

 Qiao

-0.66

blogspot

-0.65

iaries

-0.65

pan

-0.64

arthed

-0.63

 Pilgrim

-0.62

POSITIVE LOGITS

 Advertisement

0.75

oca

0.72

Brow

0.69

uke

0.65

 CLICK

0.65

oke

0.65

 ARTICLE

0.62

 notifications

0.62

 CONT

0.60

uate

0.60

Activations Density 0.148%

links or commands prompting the user to view or open content

prompts related to viewing or accessing media content

No Comments

No Known Activations