INDEX

Explanations

places or objects related to a shoreline

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "shore" and related terms

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

æł

-0.81

iator

-0.70

 cigarettes

-0.68

iggs

-0.67

ible

-0.65

ence

-0.65

isible

-0.65

ioch

-0.65

ombs

-0.64

lethal

-0.64

POSITIVE LOGITS

line

1.20

lines

1.09

 shore

1.06

side

0.97

front

0.91

yard

0.91

liner

0.90

lands

0.89

ward

0.87

bank

0.84

Activations Density 0.012%

places or objects related to a shoreline

mentions of the word "shore" and related terms

No Comments

No Known Activations