INDEX

Explanations

specific words or phrases that mention or refer to dictionaries

oai_token-act-pair · gpt-3.5-turbo

references to dictionaries or dictionary-related concepts

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

roach

-0.72

ments

-0.72

IENT

-0.70

 Elys

-0.68

 Antar

-0.67

ardy

-0.65

 hills

-0.62

rity

-0.62

psey

-0.62

mented

-0.62

POSITIVE LOGITS

 Dictionary

1.16

 dictionary

1.04

tymology

0.88

ictionary

0.86

pedia

0.86

 Britann

0.84

 diction

0.78

 definitions

0.77

initions

0.76

ãĥĥãĤ¯

0.76

Activations Density 0.017%

specific words or phrases that mention or refer to dictionaries

references to dictionaries or dictionary-related concepts

No Comments

No Known Activations