INDEX

Explanations

phrases related to puzzles or puzzle-solving

oai_token-act-pair · gpt-3.5-turbo

references to puzzles and puzzle-solving concepts

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

redits

-0.76

ankind

-0.75

umption

-0.68

ript

-0.68

lav

-0.68

ributes

-0.68

rol

-0.66

ml

-0.65

oln

-0.65

artney

-0.63

POSITIVE LOGITS

 puzzles

1.25

 puzzle

1.23

ĸļ

1.13

 Puzzle

0.97

 Puzzles

0.96

 solving

0.93

 Puzz

0.91

 puzz

0.90

 solved

0.80

 puzzling

0.78

Activations Density 0.008%

phrases related to puzzles or puzzle-solving

references to puzzles and puzzle-solving concepts

No Comments

No Known Activations