INDEX

Explanations

sequences that involve coordinating actions or plans

oai_token-act-pair · gpt-3.5-turbo

instances of the word "coordinate" and its variations

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

gery

-0.79

thur

-0.79

duc

-0.78

sburg

-0.77

adish

-0.75

gom

-0.72

den

-0.72

ther

-0.71

sn

-0.70

grim

-0.70

POSITIVE LOGITS

 coordination

0.98

 coordinating

0.98

inators

0.91

 coordinate

0.90

 coordinated

0.87

 coord

0.86

 coordinates

0.81

 teamwork

0.75

coord

0.75

 spacing

0.74

Activations Density 0.020%

sequences that involve coordinating actions or plans

instances of the word "coordinate" and its variations

No Comments

No Known Activations