INDEX

Explanations

data-driven statements or research findings

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate research findings or evidence

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

etheless

-0.85

iche

-0.83

ascus

-0.75

icultural

-0.72

plete

-0.70

ilit

-0.70

cult

-0.67

theless

-0.67

pton

-0.67

rete

-0.65

POSITIVE LOGITS

shows

0.99

manship

0.91

biz

0.88

ered

0.86

runners

0.86

heet

0.83

show

0.81

 Shows

0.80

runner

0.78

 shows

0.77

Activations Density 0.050%

data-driven statements or research findings

phrases that indicate research findings or evidence

No Comments

No Known Activations