INDEX

Explanations

short phrases related to informative content or newsletters

oai_token-act-pair · gpt-3.5-turbo

the word "Brief" indicating a focus on brief updates or summaries

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

artifacts

-0.77

coni

-0.74

CE

-0.72

OURCE

-0.68

 Malfoy

-0.67

UT

-0.65

natureconservancy

-0.64

 Scotia

-0.62

odon

-0.61

pez

-0.61

POSITIVE LOGITS

ing

1.25

gements

0.95

edly

0.92

gments

0.91

ly

0.91

ingham

0.90

0.89

edIn

0.87

ĭ

0.86

ings

0.86

Activations Density 0.027%

short phrases related to informative content or newsletters

the word "Brief" indicating a focus on brief updates or summaries

No Comments

No Known Activations