INDEX

Explanations

blog-related terms and references

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Blog" in various contexts

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

asonic

-0.72

 Scotia

-0.70

 Lauder

-0.69

TRUMP

-0.69

arching

-0.68

IELD

-0.68

 Vegas

-0.64

 dial

-0.64

inventoryQuantity

-0.62

 Winc

-0.59

POSITIVE LOGITS

gers

1.56

ging

1.36

ger

1.34

ged

1.12

glers

0.90

ges

0.90

osphere

0.89

gments

0.85

rition

0.85

post

0.85

Activations Density 0.017%

blog-related terms and references

occurrences of the word "Blog" in various contexts

No Comments

No Known Activations