INDEX

Explanations

names and locations related to historical or regional significance

oai_token-act-pair · gpt-3.5-turbo

specific place names and references to geographical locations

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

ĸļ

-0.81

 colleg

-0.78

mble

-0.74

 norm

-0.71

IGF

-0.70

 reckoning

-0.66

ITNESS

-0.66

itled

-0.65

 McMaster

-0.64

 entitle

-0.63

POSITIVE LOGITS

cius

0.83

jas

0.80

chio

0.78

illon

0.77

lang

0.77

 Blossom

0.77

igham

0.77

oglu

0.76

ESE

0.75

loe

0.75

Activations Density 0.241%

names and locations related to historical or regional significance

specific place names and references to geographical locations

No Comments

No Known Activations