INDEX

Explanations

words related to specific locations or tribes, likely the Tuareg tribe given the activations

oai_token-act-pair · gpt-3.5-turbo

the word "are" in different 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

ingen

-0.80

omez

-0.76

ured

-0.76

ulates

-0.74

uration

-0.73

ues

-0.72

enegger

-0.72

inatory

-0.71

isting

-0.69

inosaur

-0.68

POSITIVE LOGITS

tto

1.12

tta

1.08

nce

1.01

lli

1.01

llan

0.97

nda

0.93

tsky

0.92

ndra

0.92

nces

0.90

zza

0.90

Activations Density 0.018%

words related to specific locations or tribes, likely the Tuareg tribe given the activations

the word "are" in different contexts

No Comments

No Known Activations