INDEX

Explanations

keywords related to the word "Hat"

oai_token-act-pair · gpt-3.5-turbo

references to the word "Hat" and its variations in multiple contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ngth

-1.00

theless

-0.83

 glutamate

-0.75

 UNIVERS

-0.74

 confir

-0.71

sample

-0.71

hower

-0.70

successfully

-0.68

minimum

-0.68

ournal

-0.67

POSITIVE LOGITS

chet

1.05

Hat

0.98

chery

0.97

dar

0.90

Hat

0.90

red

0.82

wig

0.80

cher

0.80

ches

0.80

cot

0.74

Activations Density 0.005%

keywords related to the word "Hat"

references to the word "Hat" and its variations in multiple contexts

No Comments

No Known Activations