INDEX

Explanations

specific locations or points in a sequence

oai_token-act-pair · gpt-3.5-turbo

references to various transport stops

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

yss

-0.84

IGHTS

-0.76

ighth

-0.76

ipedia

-0.74

xual

-0.73

perty

-0.72

aths

-0.70

jri

-0.69

uth

-0.69

TPPStreamerBot

-0.69

POSITIVE LOGITS

watch

0.98

gap

0.96

oppers

0.78

opping

0.77

watching

0.73

reon

0.69

 shop

0.69

cart

0.68

ãĥ³

0.68

loading

0.67

Activations Density 0.018%

specific locations or points in a sequence

references to various transport stops

No Comments

No Known Activations