INDEX

Explanations

phrases related to summarizing and writing headlines for stories

oai_token-act-pair · gpt-3.5-turbo

the mention of significant organizations or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

eatures

-0.72

 aloud

-0.69

istical

-0.67

 Labs

-0.66

 Genius

-0.65

luster

-0.65

 Weekend

-0.64

ONSORED

-0.63

oscope

-0.63

)].

-0.62

POSITIVE LOGITS

seless

0.68

lees

0.68

elsen

0.67

awaru

0.66

 pawn

0.63

uman

0.62

Mon

0.61

ierrez

0.61

peg

0.59

hower

0.59

Activations Density 0.000%

phrases related to summarizing and writing headlines for stories

the mention of significant organizations or entities

No Comments

No Known Activations