INDEX

Explanations

time-related actions or events

oai_token-act-pair · gpt-3.5-turbo

occurrences of verbs related to actions or processes, particularly in research or survey contexts

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

wings

-0.69

roots

-0.67

 temper

-0.61

ugu

-0.61

 jungle

-0.60

 purity

-0.59

ãģĦ

-0.59

ãĤ·

-0.58

ses

-0.58

 Appears

-0.57

POSITIVE LOGITS

 Corker

0.79

 anonymously

0.75

 pseudonym

0.75

Oracle

0.74

hani

0.70

 Arpaio

0.68

herer

0.68

 electronically

0.68

nergy

0.67

actionDate

0.66

Activations Density 0.359%

time-related actions or events

occurrences of verbs related to actions or processes, particularly in research or survey contexts

No Comments

No Known Activations