INDEX

Explanations

expressions of personal beliefs or opinions

oai_token-act-pair · gpt-3.5-turbo

expressions of personal feelings and identities

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

Quotes

-0.68

Saharan

-0.66

Analysis

-0.63

eret

-0.62

Americ

-0.62

Pros

-0.62

 Gaul

-0.61

bush

-0.60

Fix

-0.59

Decre

-0.59

POSITIVE LOGITS

ãĥĥãĤ¯

0.74

 myself

0.70

ãĥ¼ãĥ«

0.69

arella

0.65

psy

0.65

 wondering

0.64

ãĤ´

0.64

aware

0.63

aze

0.63

 nervous

0.61

Activations Density 0.082%

expressions of personal beliefs or opinions

expressions of personal feelings and identities

No Comments

No Known Activations