INDEX

Explanations

references to reports or analysis in various contexts

oai_token-act-pair · gpt-3.5-turbo

repetitive conjunctions that introduce clauses or phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Ct

-0.66

rule

-0.61

chance

-0.59

zero

-0.59

 Doctrine

-0.58

animate

-0.58

mens

-0.57

Course

-0.56

ners

-0.56

addons

-0.56

POSITIVE LOGITS

 summarizes

1.20

 depicts

1.08

 accompanies

1.06

 illustrates

1.06

 contradicts

1.06

 outlines

1.05

 sheds

1.04

 contains

1.03

 compares

1.03

 culminated

1.02

Activations Density 0.204%

references to reports or analysis in various contexts

repetitive conjunctions that introduce clauses or phrases

No Comments

No Known Activations