INDEX

Explanations

names of individuals, especially related to sports and entertainment

oai_token-act-pair · gpt-3.5-turbo

names of individuals in the context of entertainment and sports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

onyms

-0.83

lihood

-0.78

abases

-0.72

yre

-0.69

elist

-0.69

ukemia

-0.69

gered

-0.68

ERAL

-0.68

liest

-0.67

roleum

-0.67

POSITIVE LOGITS

 Abrams

0.85

feld

0.78

ynski

0.75

hower

0.73

factor

0.67

="#

0.67

 Gunn

0.66

 inoc

0.66

icum

0.65

mount

0.64

Activations Density 0.012%

names of individuals, especially related to sports and entertainment

names of individuals in the context of entertainment and sports

No Comments

No Known Activations