INDEX

Explanations

names of people and organizations

oai_token-act-pair · gpt-3.5-turbo

specific names or terms related to marine life and cultural 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

 skelet

-0.70

querque

-0.69

ģĸ

-0.65

Uncommon

-0.62

CLS

-0.62

ACA

-0.62

berus

-0.60

 inval

-0.60

race

-0.59

Ì¶

-0.57

POSITIVE LOGITS

urat

0.79

ibaba

0.78

inki

0.71

ahime

0.69

igans

0.69

sung

0.69

hots

0.68

 Tanzania

0.68

gaard

0.67

oslav

0.67

Activations Density 0.862%

names of people and organizations

specific names or terms related to marine life and cultural contexts

No Comments

No Known Activations