INDEX

Explanations

proper nouns related to different locations, possibly a specific city or region

oai_token-act-pair · gpt-3.5-turbo

references to geographic locations or place names

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

ĸļ

-0.92

©¶æ

-0.74

 Uriel

-0.72

DIR

-0.68

 Pradesh

-0.67

URA

-0.67

PF

-0.65

 grapple

-0.64

 seiz

-0.63

 Expend

-0.63

POSITIVE LOGITS

itte

0.92

gger

0.89

veland

0.87

sels

0.86

olini

0.86

acket

0.84

hemy

0.82

cade

0.81

wine

0.80

cker

0.77

Activations Density 0.063%

proper nouns related to different locations, possibly a specific city or region

references to geographic locations or place names

No Comments

No Known Activations