INDEX

Explanations

numbers and numerical patterns

oai_token-act-pair · gpt-3.5-turbo

references to numerical values and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

loo

-0.79

REAM

-0.76

hips

-0.74

hops

-0.71

WARD

-0.70

RAFT

-0.68

 Denis

-0.66

ffee

-0.65

IDES

-0.64

bda

-0.64

POSITIVE LOGITS

emonic

1.10

eral

1.08

pty

0.91

quist

0.87

phys

0.84

Num

0.81

urg

0.77

BER

0.75

num

0.75

aho

0.75

Activations Density 0.017%

numbers and numerical patterns

references to numerical values and related terms

No Comments

No Known Activations