INDEX

Explanations

specific occurrences or events that happened after a particular action or event

oai_token-act-pair · gpt-3.5-turbo

phrases that relate to events or actions that occur sequentially, specifically starting with "Following."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

flo

-0.88

ice

-0.84

adj

-0.78

omet

-0.77

Ñĭ

-0.77

atron

-0.73

iman

-0.73

uci

-0.71

cia

-0.71

esse

-0.71

POSITIVE LOGITS

 revelations

0.91

 suit

0.87

 completion

0.83

 graduation

0.80

SX

0.78

 objections

0.77

 adoption

0.73

 introdu

0.71

 incidents

0.71

 passage

0.70

Activations Density 0.044%

specific occurrences or events that happened after a particular action or event

phrases that relate to events or actions that occur sequentially, specifically starting with "Following."

No Comments

No Known Activations