INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

references to individuals and their roles or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Bound

-0.67

 Affordable

-0.64

Untitled

-0.62

reach

-0.62

interstitial

-0.61

AGE

-0.61

Kit

-0.60

ulp

-0.60

Nec

-0.60

 Guys

-0.59

POSITIVE LOGITS

 oversaw

1.01

 owns

1.01

 oversees

0.97

 resided

0.97

 accompanies

0.95

 preceded

0.94

oping

0.90

 accompanied

0.90

 presided

0.89

 died

0.89

Activations Density 0.101%

names of individuals

references to individuals and their roles or actions

No Comments

No Known Activations