INDEX

Explanations

The neuron is looking for occurrences of the word "while"

oai_token-act-pair · gpt-3.5-turbo

the word "while" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

emb

-0.70

 backs

-0.69

 arranged

-0.67

 iron

-0.66

 fields

-0.66

 organized

-0.66

 organised

-0.65

 legend

-0.65

 Ireland

-0.65

 coming

-0.65

POSITIVE LOGITS

while

3.67

theless

1.17

during

1.13

instead

1.09

until

1.05

continue

1.04

soDeliveryDate

1.03

despite

1.01

lest

1.00

before

1.00

Activations Density 0.022%

The neuron is looking for occurrences of the word "while"

the word "while" in various contexts

No Comments

No Known Activations