INDEX

Explanations

terms related to the concept of "forward"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "forward" 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

colo

-0.89

chens

-0.81

regon

-0.77

azo

-0.76

enez

-0.75

ENE

-0.74

abase

-0.71

gins

-0.71

lez

-0.70

mingham

-0.69

POSITIVE LOGITS

forward

0.84

 thinker

0.78

 forwarding

0.75

olicy

0.71

 characterization

0.71

 thinkers

0.71

 forward

0.70

wards

0.70

ward

0.69

Iterator

0.67

Activations Density 0.017%

terms related to the concept of "forward"

occurrences of the word "forward" in various contexts

No Comments

No Known Activations