INDEX

Explanations

years and time-related terms

oai_token-act-pair · gpt-3.5-turbo

references to time indicators, specifically years and days

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

OTUS

-0.73

":["

-0.72

rity

-0.70

 RELE

-0.70

untled

-0.67

ategory

-0.66

ACTED

-0.66

agonist

-0.64

uckland

-0.64

 constitu

-0.63

POSITIVE LOGITS

dream

0.98

ning

0.89

table

0.89

worms

0.85

long

0.84

days

0.84

wise

0.82

lies

0.80

mar

0.77

books

0.77

Activations Density 0.111%

years and time-related terms

references to time indicators, specifically years and days

No Comments

No Known Activations