INDEX

Explanations

words related to events or occurrences

oai_token-act-pair · gpt-3.5-turbo

variations of the word "happen."

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

ordes

-0.80

ilts

-0.74

ription

-0.73

ulton

-0.72

IPM

-0.72

rawn

-0.69

orde

-0.68

rongh

-0.67

onga

-0.67

iyah

-0.67

POSITIVE LOGITS

lihood

0.99

 Happ

0.90

hof

0.84

LY

0.78

çĲ

0.78

 Procedure

0.76

ãĤ´ãĥ³

0.76

 happening

0.75

 Alive

0.75

lander

0.75

Activations Density 0.055%

words related to events or occurrences

variations of the word "happen."

No Comments

No Known Activations