INDEX

Explanations

mentions of technology-related terms and information, such as operating systems, models, revisions, and network codes

oai_token-act-pair · gpt-3.5-turbo

terms related to technical specifications and classifications

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ãģį

-0.72

isks

-0.70

ilus

-0.70

eln

-0.69

assic

-0.69

etsk

-0.68

azy

-0.68

owan

-0.68

mobi

-0.67

vind

-0.67

POSITIVE LOGITS

 selector

1.00

 affiliation

0.93

 chosen

0.84

 preference

0.83

 choice

0.77

names

0.75

etting

0.73

 specified

0.72

 Specific

0.71

specified

0.71

Activations Density 0.380%

mentions of technology-related terms and information, such as operating systems, models, revisions, and network codes

terms related to technical specifications and classifications

No Comments

No Known Activations