INDEX

Explanations

the word "Senator" followed by a person's name

oai_token-act-pair · gpt-3.5-turbo

the word "Senator" and its associated instances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 upkeep

-0.69

illusion

-0.66

phal

-0.66

 Britann

-0.66

footed

-0.65

lay

-0.63

 polic

-0.61

 curfew

-0.61

lot

-0.60

 showc

-0.60

POSITIVE LOGITS

ial

1.05

 Grassley

1.02

 Dianne

1.02

 McCain

0.99

 Schumer

0.98

 Bernie

0.91

 Senator

0.89

ileaks

0.89

arians

0.88

 filib

0.88

Activations Density 0.026%

the word "Senator" followed by a person's name

the word "Senator" and its associated instances

No Comments

No Known Activations