INDEX

Explanations

specific years from the 1950s, 1960s, and 1970s

oai_token-act-pair · gpt-3.5-turbo

references to specific decades and their associated years

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

abol

-0.73

head

-0.71

atro

-0.70

plan

-0.70

ONS

-0.70

henko

-0.68

abolic

-0.68

spons

-0.67

que

-0.67

ques

-0.64

POSITIVE LOGITS

1.29

sburg

1.05

sie

0.98

sburgh

0.96

enthal

0.84

eenth

0.84

si

0.78

sd

0.77

enstein

0.77

ies

0.76

Activations Density 0.017%

specific years from the 1950s, 1960s, and 1970s

references to specific decades and their associated years

No Comments

No Known Activations