INDEX

Explanations

mentions of the word "Hill"

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of the term "Hill," specifically in varying contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Alonso

-0.75

lihood

-0.69

============

-0.67

 Bucc

-0.67

âĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪ

-0.66

 Guerrero

-0.65

 oppos

-0.64

uality

-0.63

glomer

-0.63

 Norn

-0.62

POSITIVE LOGITS

iard

1.50

tops

1.15

top

1.14

side

1.13

yer

1.05

bill

1.04

marks

0.97

shire

0.97

arious

0.92

ibur

0.92

Activations Density 0.047%

mentions of the word "Hill"

the repeated mention of the term "Hill," specifically in varying contexts

No Comments

No Known Activations